ArticlePDF AvailableLiterature Review

Abstract

The aim of this review is to consider variation in mating p among females. We define mating p as the sensory and behavioural properties that influence the propensity of individuals to mate with certain phenotypes. Two properties of mating p can be distinguished: (i) ‘preference functions’–the order with which an individual ranks prospective mates and (2)‘choosiness’ -the effort an individual is prepared to invest in mate assessment. Patterns of mate choices can be altered by changing the costs of choosiness without altering the preference function. We discuss why it is important to study variation in female mating behaviour and identify five main areas of interest: Variation in mating p and costs of choosiness could (i) influence the rate and direction of evolution by sexual selection, (2) provide information about the evolutionary history of female p, (3) help explain inter-specific differences in the evolution of secondary sexual characteristics, (4) provide information about the level of benefits gained from mate choice, (5) provide information about the underlying mechanisms of mate choice. Variation in mate choice could be due to variability in preference functions, degree of choosiness, or both, and may arise due to genetic differences, developmental trajectories or proximate environmental factors. We review the evidence for genetic variation from genetic studies of heritability and also from data on the repeatability of mate-choice decisions (which can provide information about the upper limits to heritability). There can be problems in interpreting patterns of mate choice in terms of variation in mating p and we illustrate two main points. First, some factors can lead to mate choice patterns that mimic heritable variation in p and secondly other factors may obscure heritable p. These factors are divided into three overlapping classes, environmental, social and the effect of the female phenotype. The environmental factors discussed include predation risk and the costs of sampling; the social factors discussed include the effect of male–male interactions as well as female competition. We review the literature which presents data on how females sample males and discuss the number of cues females use. We conclude that sexual-selection studies have paid far less attention to variation among females than to variation among males, and that there is still much to learn about how females choose males and why different females make different choices. We suggest a number of possible lines for future research.
Biol.Rev.(), ,pp.
Printed in Great Britain

VARIATION IN MATE CHOICE AND MATING PREFERENCES:
A REVIEW OF CAUSES AND CONSEQUENCES
BMICHAEL D. JENNIONS" MARION PETRIE#
Animal Behaviour Research Group,Department of Zoology,University of Oxford,
South Parks Road,Oxford OXPS,UK
(Received  November ;revised  July ;accepted  July )
ABSTRACT
The aim of this review is to consider variation in mating preferences among females. We
define mating preferences as the sensory and behavioural properties that influence the
propensity of individuals to mate with certain phenotypes. Two properties of mating
preferences can be distinguished: () ‘preference functions’ the order with which an
individual ranks prospective mates and () ‘choosiness’ – the effort an individual is prepared
to invest in mate assessment. Patterns of mate choices can be altered by changing the costs of
choosiness without altering the preference function. We discuss why it is important to study
variation in female mating behaviour and identify five main areas of interest: Variation in
mating preferences and costs of choosiness could () influence the rate and direction of
evolution by sexual selection, () provide information about the evolutionary history of female
preferences, () help explain inter-specific differences in the evolution of secondary sexual
characteristics, () provide information about the level of benefits gained from mate choice, ()
provide information about the underlying mechanisms of mate choice. Variation in mate
choice could be due to variability in preference functions, degree of choosiness, or both, and
may arise due to genetic differences, developmental trajectories or proximate environmental
factors. We review the evidence for genetic variation from genetic studies of heritability and
also from data on the repeatability of mate-choice decisions (which can provide information
about the upper limits to heritability). There can be problems in interpreting patterns of
mate choice in terms of variation in mating preferences and we illustrate two main points.
First, some factors can lead to mate choice patterns that mimic heritable variation in
preferences and secondly other factors may obscure heritable preferences. These factors are
divided into three overlapping classes, environmental, social and the effect of the female
phenotype. The environmental factors discussed include predation risk and the costs of
sampling; the social factors discussed include the effect of male–male interactions as well as
female competition. We review the literature which presents data on how females sample
males and discuss the number of cues females use. We conclude that sexual-selection studies
have paid far less attention to variation among females than to variation among males, and that
there is still much to learn about how females choose males and why different females make
different choices. We suggest a number of possible lines for future research.
Key words : Female mate choice, mate preferences, mating decisions, variation, sexual selection,
mate sampling, female–female competition.
"Present address: Smithsonian Tropical Research Institute, P.O. Box , Balboa, Ancon, Republic of
Panama.
#Present address; Evolution and Behaviour Research Group, Ridley Building, University of Newcastle,
Newcastle-upon-Tyne NERU, UK.
 M. D. J  M. P
CONTENTS
I. Introduction .............. 
() Historic background ............ 
() Definitions ............. 
II. Why study variability in female mating preferences and mate choice ? . . . 
III. Explaining variation in mate choice decisions ........ 
IV. Genetic variation in mating preferences ......... 
() Genetic studies of heritability .......... 
() Repeatability as an indicator of heritability ....... 
V. Non-heritable causes of variation in mate-choice decisions ..... 
VI. Environmental factors ............ 
() Time and energy costs of sampling ......... 
() Predation risk ............. 
() Territory or resource quality .......... 
() Environmental effects on signal detection and discrimination .... 
VII. Social factors : the effects of conspecifics ......... 
() Interactions between males .......... 
() Variability in male phenotypes .......... 
() Female–female competition .......... 
() Female mate copying ........... 
() Density and the operational sex ratio ........ 
VIII. Female phenotypes ............. 
IX. How do females choose males ? .......... 
() Sampling tactics ............ 
() How many cues do females use? ......... 
X. Conclusions .............. 
XI. Acknowledgements ............. 
XII. References .............. 
I. INTRODUCTION
()Historic background
The evolution of extravagant secondary sexual characters (‘ ornaments’) has been the
subject of intense theoretical and empirical investigation (Andersson, ). However,
in spite of the enormous amount of work conducted, a few issues have dominated to the
exclusion of others. Initially, Darwin’s claim that female choice played a role in the
evolution of male secondary sexual characters was doubted (Cronin, ). Early
research therefore aimed at demonstrating this phenomenon by looking at male mating
success. Was variation in male mating success non-random? If so, could it be related
to differences in male traits ? Numerous correlational and experimental studies from
many taxa now confirm that males with increased ornamentation, or possessing certain
attributes, have a mating advantage arising from female mate choice (reviews:
Bradbury & Andersson,  ; Ryan & Keddy-Hector, ; Andersson, ; Møller,
a; Johnstone, ).
Subsequent research then concentrated on determining what selection pressures
could maintain seemingly costly female mating preferences. Three main explanations
have been proposed : () preferences may be directly selected due to direct benefits
which increase female survival or fecundity (Reynolds & Gross, ); () preferences
may be maintained by indirect selection due to genetic benefits that increase offspring
fitness. Two types of genetic benefits are recognized. Choice may lead either to the
production of offspring with genotypes that increase viability (‘good genes’ or ‘ viability
Variation in mate choice: a review 
genes’), and}or to the production of sons with genotypes that make them more
attractive (Fisherian traits) (reviews: Kirkpatrick & Ryan, ; Andersson, ); ()
preferences may be maintained as pleiotropic effects of natural selection on female
sensory systems in contexts other than mate choice, such as foraging or predator evasion
(Enquist & Arak, ; Ryan, ; Arak & Enquist, ). If benefits gained outside
the context of mating outweigh the cost of preferences during mate choice, net selection
will maintain preferences (Christy, ). However, if the mating preference is costly,
there may be selection for modifiers that alter sensory functioning or processing during
mate choice, thereby altering the preference (Christy & Backwell, ).
Despite interest in the question of what maintains female preferences, most work has
focused on male traits. First, significant heritability of preferred male traits has been
documented (Hedrick,  ; Hill, ; Norris, ; Pomiankowski & Møller, ).
Second, several studies indicate that preferred males sire offspring with higher viability
or fecundity (e.g. Reynolds & Gross, ; Norris, ; Petrie, ). Thus, a current
question of great interest is how ornaments honestly signal ‘ genetic ’ quality. This has
led to renewed interest in the handicap principle (Zahavi,  ; Grafen, ), and has
encouraged research into the costs of ornamentation (Evans & Thomas, ;
Balmford, Thomas & Jones,  a; Borgia,  ; Jennions,  ; Møller & de Lope,
). There have also been several attempts to correlate degree of ornamentation with
possible indices of male ‘ quality’ (Møller & Pomiankowski, a,b; for a detailed
review see Johnstone, ).
Much attention has been given to the often larger amounts of phenotypic variation
in the size of male ornaments compared with other morphological traits (Alatalo,
Ho
$glund & Lundberg,  b; Barnard, ). One popular explanation for this trend
is that variation in male quality determines the costs incurred with increased investment
in ornaments, leading to condition-dependent expression of sexual traits in accordance
with the handicap principle of signalling (Grafen, ). In contrast, few researchers
attempt to account for variation in female mating preferences in terms of differing costs
and benefits. Males appear to trade-off the benefits of larger ornaments against the
increased costs of developing and maintaining them (Møller & de Lope, ).
Likewise, we might also expect females to trade-off the benefits gained from being
choosy against the costs of stronger preferences. Moreover, because of differences in
age, experience, body condition and size, we might also expect that these trade-offs will
differ among females (Ho
$glund & Alatalo, ). Variation in female quality should
determine the point at which investment in mate choice is optimized, leading to
condition-dependent expression of preferences. Although we now know something
about the benefits of mate choice for both direct (e.g. Hill, ) and indirect benefits
(e.g. Norris, ; Petrie, ), we know almost nothing about the costs. Do different
preference functions carry different costs? Does sampling increase the risk of predation
or disease transfer ? How much time and energy is invested into sampling males ? What
tactics can females use to reduce the potential costs of sampling?
Recently, it has been suggested that a typological view prevails in which female
mating preferences are seen as species-specific, stereotypic traits (W. Eberhard & W.
Wcislo, in preparation). It is claimed that the implicit view is along the lines of ‘in
species X, females prefer to mate with males with larger ornaments, and this preference
is reflected in the pattern of mate choice’. With a few notable exceptions, there has been
 M. D. J  M. P
little research into: phenotypic variation in mating preferences; female sampling
tactics; analysis of constraints on optimal mate choice ; variation among individuals in
their mate-choice decisions; or consideration of the consequences of this variability for
evolution by sexual selection. In this article, we aim to consider why it is important to
study variation in female mating behaviour, and then to review our current
understanding of possible causes.
()Definitions
Following Heisler et al.(), we define ‘mating preferences’ as the sensory and
behavioural properties that influence the propensity of individuals to mate with certain
phenotypes. ‘Mate choice’ is then defined as the pattern of mating which arises, in part,
because of these mating preferences. We further subdivide ‘mating preferences
because there are two properties that can be distinguished conceptually and, more
importantly, sometimes empirically. We define ‘preference functions’ as the order in
which an individual ranks prospective mates ceteris paribus; and ‘choosiness’ as the
effort or energy that an individual is prepared to invest in assessing mates, both in terms
of the number of mates sampled and the amount of time spent examining each mate.
Increased choosiness could arise by changes in sampling tactics, or by a higher mating
threshold. We believe the distinction between preference function and choosiness is
useful. The pattern of mate choice can be altered by changing the costs of female
choosiness with no apparent change in terms of which a male a female would most
prefer to mate with (i.e. an unchanged preference function) (e.g. Milinski & Bakker,
; Hedrick & Dill, ; Backwell & Passmore, in press). Our subdivision also
accords with common usage: ‘prefer implies preconceived partiality for one thing over
another but does not always connotate the actual getting of what one chooses ’ (Webster’s
Dictionary).
II. WHY STUDY VARIABILITY IN FEMALE MATING PREFERENCES AND MATE CHOICE ?
There are many reasons why variability in female preferences is worth studying.
Here, we provide an incomplete list, mentioning five areas of research which we find
particularly interesting.
First, variation in mating preferences and costs of choosiness influences the rate and
direction of evolution by sexual selection. Increased variability in preferences which
affects the mean level of a preference will decrease the intensity of female-driven
directional selection on male ornaments, as well as subsequent indirect selection on the
preference itself. Changes in the frequency distribution of mating preference functions
may also lead to shifts in the net direction of selection and, in some cases, disruptive
selection may even occur, leading to speciation (Turner & Burrows, ). In contrast,
a sampling tactic such as mate copying increases the advantages of common male traits,
making it more difficult for novel traits to evolve (Kirkpatrick & Dugatkin, ). Costs
to choosiness also decrease the intensity of sexual selection. Gomulkiewicz ()
modelled the effect of limited female choice by adjusting the average number of males
that females sampled. Decreased sampling made it more difficult for novel male traits
to evolve. When there is individual variation in susceptibility to costs, however, this
may cause shifts in the direction of selection. For example, consider a species where
Variation in mate choice: a review 
some females prefer shorter-tailed males and others longer-tailed males. If constraints
on choosiness differ between the two types of females and vary over time or space, or
are subject to frequency-dependent selection, fluctuations in the direction of selection
will result.
A key variable in recent models of the evolution of multiple male sexual traits is the
cost of mating preferences. For Fisherian traits, increased costs (greater than the
additive cost of each individual preference) may still lead to the evolution of multiple
preferences (Pomiankowski & Iwasa, ). However, for traits indicating viability,
only a preference for a single trait is stable if assessment of additional traits increases
costs disproportionately (Iwasa & Pomiankowski, ). Similarly, a model which
predicts continual changes in mating preferences and repeated evolution of male traits
assumes that stronger preferences carry greater costs (Iwasa & Pomiankowski, ). In
addition, the model only predicts repeated evolution of new preferences when the
correlation between trait and preference exceeds a threshold value. This value is
strongly influenced by the intensity of stabilizing natural selection on the preference.
Kirkpatrick & Barton () suggest that the threshold is unlikely to be reached under
natural conditions. Moreover, the effect of variation in female quality has not even been
incorporated into these models. Clearly, however, the usefulness of these models
depends crucially on quantitative data on the costs of mate choice. How strong is
stabilizing natural selection on preferences ? Is a preference for larger male ornaments
more costly? Do costs increase disproportionately when females assess several male
traits?
Second, variability in mating preferences provides information on their evolutionary
history. Specifically, Fisherian evolution of ornaments predicts considerable heritable
variation in female mating preferences, both within and between populations (Lande,
). At present, there is comparatively little information on the heritability or
coefficient of additive genetic variation (cf. Pomiankowski & Møller, ) of mating
preferences (but see Bakker & Pomiankowski, ). Knowledge of genetic differences
between populations may also provide data on differences in selection regimes between
areas and the extent to which preferences are subject to natural selection (but see
Houde, ). The genetic basis of female preferences may also provide some insight
into the speciation process. Turner & Burrows () suggested that the genetic bases
of preferences may lead to different speciation rates among lineages. In contrast to
Fisherian models, female choice for males with ‘viability genes’ requires neither
genotypic nor phenotypic variation among females (e.g. Grafen, ); although
polygenic models obviously assume a heritable basis to preferences, and generally
predict coevolution of preference and preferred trait (Bakker, ).
Regardless of whether Fisherian or viability-gene processes operate, a positive
genetic correlation between preference and male trait indicates that these traits have
coevolved. Sensory exploitation is often presented as the main alternative to models
requiring genetic coevolution (Ryan, ). Current phylogenetic tests of sensory
exploitation are based on female preferences remaining stable through speciation events
(Ryan, ; Shaw, ). The implicit suggestion is that a single, invariant female
preference shared by ancestral and derived species can be characterized (Christy &
Backwell, ). However, if there is population variability in female preferences,
stability not only requires that the mean preference remains the same, but also that the
 M. D. J  M. P
frequency distribution of preference functions and choosiness in the population is
unchanged. At present, evidence for stability is based on data that the mean preference
function at the population level is unchanged (e.g. Ryan et al., ). However,
selection on female choosiness could lead to very different patterns of mate choice. For
example, two species could both have auditory tuning curves with identical best
excitatory frequencies. The effective mating preferences will be very different,
however, if females in one species sample several males, while in the other they mate
with the first male encountered.
Third, it may be possible to explain inter-specific differences in the evolution of
secondary sexual characters by relating different tactics of mate choice to ecological,
social or morphological factors. To date, most attempts to explain variation in
ornamentation among species invoke strong natural selection against the elaboration of
male traits (e.g. Balmford et al.,  a; Winquist & Lemon, ). However, both the
potential for females to be choosy and variability in preference functions may be equally
important predictors of differences in male ornamentation. Factors that inhibit female
choice will reduce selection for elaborate male traits. Variation in the opportunity for
mate choice may also affect other features of mating behaviour. For example, Slagsvold
et al.() suggest that polygyny in pied flycatchers (Ficedula hypoleuca) is partly due
to high female sampling costs leading to limited searches for unpaired males. (For a
general discussion of the costs of choosiness see Noe
$& Hammerstein, .) Sullivan
() has also noted that, given severe time constraints on female choice, females are
likely to use static morphological traits that are quickly assessed. When more time is
available, choice may be based on behavioural displays which take longer to perform.
Hence, information on the duration over which females assess males may explain some
inter-specific variation in male ornamentation.
The psychophysics of female choice may also explain some variation in male
ornamentation. Cohen () has noted that Weber’s ‘ law’ provides an inevitable
constraint on male ornamentation. This ‘law’ summarizes a general finding from
psychology that the ability of an individual to discriminate a fixed size difference
between stimuli decreases as the absolute size of the stimuli increases. It may therefore
be more difficult for females to distinguish between males as the size of ornaments
increases, reducing the benefits of choosiness. Furthermore, the ability to distinguish
between stimuli may vary among taxa and different stimuli (e.g. acoustic versus visual)
which could also generate inter-specific differences. Theoretical models often make
assumptions about female discriminatory abilities, which need to be tested. For
example, Iwasa & Pomiankowski () state that ‘small ornaments are worse
indicators of male viability because differences in size are more difficult to distinguish’.
However, Weber’s law suggests that the function relating efficiency as an indicator to
signal size is more likely to be a parabola than a continually increasing function.
Fourth, given phenotypic plasticity in female mating preferences, it is possible to
manipulate the costs of choosiness and examine the effect on mate choice. This should
provide information about the benefits associated with discriminatory mating. For
example, if benefits are small, we should be able to detect cost-reducing changes in
female sampling behaviour. Experimentally, costs can often be precisely controlled,
and the benefits conferred by choosing can then be assessed by titrating known costs
against unknown benefits. It is intriguing, for example, that a cost to a female as
Variation in mate choice: a review 
seemingly small as swimming against a water current can lead to a reduction in mating
success for redder bellied male three-spined sticklebacks (Gasterosteus aculeatus)
(Milinski & Bakker, ). This implies that the benefits of mating with redder males
are fairly small, even if male redness does signal some aspect of male quality (Milinski
& Bakker, ).
Fifth, once the importance of variability in female mating preferences is recognized,
a more mechanistic account of mate choice should emerge. Only when traits show
variability can one look for correlations which eventually facilitate causal explanations.
Other areas in behavioural ecology have benefited greatly by integrating functional and
mechanistic approaches (Real, ). Among-female variability should provide
information on the neurobiology and psychology of female choice (W. Eberhard & W.
Wcislo, in preparation), and the interaction between sensory systems and ecological
conditions. As Ryan () has noted, knowledge of mechanisms provides a stronger
base when explaining why certain traits have evolved (e.g. Haines & Gould, ). In
the same way that researchers have been challenged to explain variability in male
ornamentation, they will also need to account for variability among females. There is
now evidence that the interaction between female condition and investment in sensory
apparatus, or time and energy allocated to mate choice, provides part of the explanation
(see Section VIII). In general, we are missing valuable opportunities to explain
variation in data, which is presently dismissed as ‘statistical noise’. Empirical data takes
time to collect, and should be used as fully as possible. For example, in two-choice
mating experiments, statistical analysis is often limited to binomial tests determining
whether there is a significant preference for a trait at the population level. But why do
some females choose the ‘ non-preferred’ trait? Numerous explanations exist but these
are rarely explored. Females may simply make mistakes (Wiley, ), vary in their
receptivity (Rowland et al., ), their ability to discriminate stimuli (Cohen, ;
Gerhardt, ), their choosiness or in their preference functions. For example,
Jennions, Backwell & Passmore (), Polakow et al.() and Ryan, Perrill &
Wilczynski () found that female body size explained some variation in choice in
phonotaxis experiments using frog calls that differed in spectral frequencies. In the
cricket frog Acris crepitans, body size appears to be related to preference functions,
while in the reed frog Hyperolius marmoratus it may influence the ability of females to
distinguish stimuli.
III. EXPLAINING VARIATION IN MATE-CHOICE DECISIONS
When a group of females are presented with the same choice of mates, not all females
choose to mate with the same male. For example, in lekking species, mating is not
confined to a single ‘preferred’ male (reviewed by Balmford, ; Wiley,  ;
Ho
$glund & Alatalo, ). Why? An obvious source of variation in mate choice is that
females vary in their mating preferences. This could be due to variability in preference
functions, degree of choosiness, or both. This variation may arise due to genetic
differences, developmental trajectories, or proximate environmental factors.
Heritable variation among individuals is the raw material necessary if evolution is to
occur. It is therefore crucial that we understand why there is variability in mating
preferences. Our aim in this review is to highlight possible causes of variability, and to
consider the consequences for evolution by sexual selection. Reflecting our interests as
 M. D. J  M. P
field biologists, we have tried to provide real-life examples, and ideas that can be
empirically tested. A genuine gap in our knowledge exists, but one which is amenable
to investigation.
IV. GENETIC VARIATION IN MATING PREFERENCES
Most theoretical models of the evolution of female choice assume a heritable genetic
basis to mating preferences (e.g. Fisher, ; Lande, ; Iwasa, Pomiankowski &
Nee, ). Studies which demonstrate heritable variation are thus central to their
validation (Boake, ; Bakker, ; Ritchie, a). At equilibrium in the Fisherian
process, and in most polygenic models of ‘viability gene’ sexual selection, theory
predicts positive genetic covariance between the mating preference and male ornament
(Bakker, ; Breden, Gerhardt & Butlin, ; Pomiankowski & Sheridan, ).
Evidence for genetic variation in mating preferences has recently been reviewed in
detail by Butlin () and Bakker & Pomiankowski (). Here, we provide a brief
review of evidence obtained from controlled breeding experiments (selection studies or
lineage analysis) and examination of discrete genetic analyses. We then focus in detail
on the use of data on repeatability of mate choice. In many species, breeding
experiments are impossible but repeated testing of mate choice can provide information
on the upper limits to heritability. Perhaps more interestingly, it also increases the
likelihood of detecting relationships between mate-choice decisions and contingent
environmental or phenotypic factors.
()Genetic studies of heritability
Several experiments have tested for heritable mating preferences by selecting for
specific mating preferences. In one of the first studies by Majerus, O’Donald & Weir
(), female two-spotted ladybird (Adalia bipunctata) that mated with melanistic
males were selected for each generation. The population showed a significant increase
in the proportion of females mating with melanistic males, indicating a heritable basis
to mating preferences. Moreover, the pattern of mate choice was consistent with
variation in this mating preference being due to only one or a few genes (Majerus et al.,
). Unfortunately, recent work shows that the interpretation of these results is less
clear cut (reviewed by Ritchie,  a). Attempts to repeat the earlier experiments and
select for mating preferences for melanistic males, both from wild stock and from
lineages from earlier experiments, failed (Kearns et al., ). O’Donald & Majerus
() have highlighted some potential problems with the study of Kearns et al.()
and also present some new data which are suggestive of a genetic basis to the preference.
Other selection studies have also revealed additive genetic variation in mating
preferences in a range of species [guppies, Poecilia reticulata (Houde,  ; but see
Breden & Hornaday, ) ; stalk-eyed flies, Cyrtodiopsis dalmanni (Wilkinson & Reillo,
); fruitflies, Drosophila melanogaster (e.g. Crossley, ; Kaneshiro, ), D.
mercatorum (Ikeda & Mauro, ), D.mojavensis (Koepfer, ) ; bollworms,
Pectinophora gossypiella (references in Collins & Carde
!,); planthoppers,
Ribautodelphax imitans (De Winter, ); and grasshoppers, Chorthippus brunneus
(Charalambous, Butlin & Hewitt, )]. In these studies, the mating preference was
either directly selected for, or the preferred trait was selected for and the preference
changed as a correlated response (see Bakker & Pomiankowski, ).
Variation in mate choice: a review 
Discrete genetic effects are associated with different mating preferences in several
species. In the sulphur butterfly Colias eurytheme, there are two female colour morphs
which differ in their choice of mates. Female coloration is heritable (Sappington &
Taylor, ). Tebb & Thoday () found that female fruitflies, D.melanogaster,
differing in genotype (heterozygote versus the two homozygotes) preferred different
male genotypes. Similarly, Heisler () presented evidence for a genetic basis to
female mating preferences for wild-type and yellow mutant males in D.melanogaster.
In seaweed flies Coelopa frigida, different female preferences are associated with
different inversion karyotypes (Gilburn, Foster & Day,  ; Gilburn & Day, )
and different alleles at the alcohol dehydrogenase locus (Engelhard, Foster & Day,
). Some breeding studies using lineage analysis (e.g. father–daughter, full-
sib}half-sib design) also suggest a genetic basis to preferences. This has been shown by
father–daughter analysis of a pheromone-based mating preference in the cockroach
Nauphoetia cinerea (Moore, ). This finding is interesting as pheromones appear to
signal dominance and active choice of dominant males leads to both indirect and direct
benefits (Moore, ). This begs the question why all females do not prefer dominant
males to the same extent? Father-son analysis in a moth (Argyrotaenia velutinana) also
indicates a heritable male mating preference for female pheromones (Roelofs et al.,
). In brown planthoppers (Nilaparvata lugens), work on repeatability of female
responses to male acoustic signals showed differences between isofemale lines suggestive
of a genetic basis. However, the variation in female responses was in the ‘preference
window’ (range of stimuli responded to), not in the mean preference (Butlin, ). Of
course, not all studies of the underlying basis of variation in preferences reveal a genetic
component. For example, Johnson et al.() found no evidence from
mother–daughter analysis for heritable mating preferences in red junglefowl (Gallus
gallus). Likewise, Nicoletto () found no heritable female mating preference for
amount of male orange coloration in guppies (P.reticulata). In a review, Ritchie
(a) also cited two unpublished studies in which selection for female preferences
was unsuccessful (see also Boake, ; Ritchie, b). As always, the possibility that
negative results are less likely to be published should be considered.
Bakker (), using a full-sib}half-sib breeding design, reported a genetic basis to
female preferences for male redness in three-spined stickleback (G.aculeatus).
Moreover, there was also a positive genetic correlation between male colouration and
female preference. Father–daughter analysis also suggested a positive genetic
correlation between preference and preferred trait in the cockroach, N.cinerea (Moore,
,) and the redbanded leafroller moth, A.velutinana (Roelofs et al., ).
Several recent studies have applied artificial directional selection for preferred traits
and recorded correlated changes in preferences. When these changes are in the same
direction as that of the selection on the male trait, this indicates a positive genetic
correlation [stalk-eyed fly, C.dalmanni (Wilkinson & Reillo, ); guppy, P.reticulata
(Houde, ; but see Breden & Hornaday, ) ; fruitfly, D.mercatorum (Ikeda &
Mauro, ); bollworms, P.gossypiella (several studies, see Collins & Carde
!,);
planthopper, R.imitans (De Winter, ); and the grasshopper, C.brunneus
(Charalambous et al., )]. These results are intriguing, because some theoretical
models suggest that genetic correlations are unlikely to be maintained by non-random
mating unless effective population size is very large (Nichols & Butlin,  ; see
 M. D. J  M. P
Gilburn & Day, ). For a general critique of these studies see Breden et al.()
and Pomiankowski & Sheridan ().
Studies looking at differences between populations can also provide information on
the genetics of mating preferences. For example, research on guppies (P.reticulata)
(Houde, ; Houde & Endler, ; Endler & Houde, ; cf. Nicoletto, ) and
grasshoppers (Ephippiger ephippiger) (Ritchie, ) both show heritable variation in
mating preferences. In cricket frogs (A.crepitans), differences among populations in
female auditory tuning curves exist. These are independent of mean population body
size, although other environmental factors that differ between populations cannot be
excluded (Ryan & Wilczynski, ; Ryan et al., ). Demonstrating a genetic basis
to preferences does not, however, address the difficult question of the maintenance of
within-population variation. There are also difficulties in accounting for differences
among populations in terms of the main models for the evolution of female preferences
(see Houde, ). However, the apparent absence of among-population variation (at
least as characterized by ‘ mean preferences’), despite variation in preferred male traits,
suggest that there are adaptive or genetic constraints on the evolution of preferences
and}or ornaments (Ryan & Wagner,  ; Ryan et al., ). Constraints on preference
evolution are consistent with some sensory exploitation models for the evolution of
male ornaments (Shaw, ; but see Christy, ; Christy & Backwell, ).
()Repeatability as an indicator of heritability
With large, long-lived animals, the practicality of performing breeding experiments
is daunting (but see Johnson et al., ). In these species, investigating the
repeatability of mate choice (whether or not individuals make the same choice when
reassessing the same potential mates) offers another approach to the study of variation
(Boake, ; Gerhardt, ). Repeatability provides an index of the level of
phenotypic variation among females and sets an upper limit to heritability. For
example, if females do not show repeatable mate-choice decisions, there is presumably
no variation in mating preferences among females (but see below). Observed variation
in mate choice can therefore be attributed to within-individual variation or stochastic
events. Conversely, repeatable patterns of mate choice that vary among females suggest
individual differences in mating preferences.
Few experimental studies of mate-choice repeatability have been performed.
Gerhardt () found that nine out of  female treefrogs (Hyla cinerea) chose the
same stimulus in at least % of tests, when offered a choice between a conspecific and
hybrid call. Robertson () noted that when female red-groined toadlets (Uperolia
laevigata) were experimentally separated from their mate and then released back into
the chorus they tended to remate with the same male. Møller ( b) reported
repeatable female mate choice for male tail length in the barn swallow (Hirundo rustica),
and Godin & Dugatkin () found repeatable female choice for amount of orange
colouration in guppies (P.reticulata). Wagner, Murray & Cade () also reported
repeatable female responses with respect to some parameters of advertisement calls in
field crickets (Gryllus integer). In a two-choice laboratory experiment, Poulin ()
also found that bullies (Gobiomorphus breviceps) preferentially chose to mate with the
male they first visited.
In contrast, Ritchie  b) did not find repeatable female preferences for calls
Variation in mate choice: a review 
varying in syllable number within a population of bushcrickets (E.ephippiger), and
Boake () found no repeatability of female choice based on pheromones in a
mealworm beetle (Tenebrio castaneum). Perhaps the most interesting negative result is
that of Ligon & Zwartjes (a). They offered female red junglefowl (G.gallus) five
successive opportunities to choose between two males. In their initial choice,  out of
 females chose a male with a large ornamental head comb; but in the subsequent four
trials only four of the females mated exclusively with the large-combed male. Thus,
while the females appeared to be able to discriminate between males, most of them
actively chose to mate with both males. There is now evidence that females sometimes
mate multiply to generate sperm competition (e.g. Madsen et al., ), and the
implications of this for Fisherian and ‘viability gene’ models for mating preferences for
males with large ornaments have yet to be fully explored. Superficially, there is a
contradiction between the claim that females choose males with large ornaments
because this indicates possession of ‘good genes’ and the claim that females mate with
a variety of males (including males with small ornaments) because this improves
offspring viability. Olsson & Madsen () suggest that multiple mating in lizards may
occur because males do not possess phenotypic traits that are reliable indicators of
genetic quality. Females may therefore promote sperm competition as the only way in
which to obtain sperm from high-quality males. In contrast, Stockley et al.()
suggest that multiple mating in common shrews, Sorex araneus, is a mechanism to
reduce the risk of inbreeding. The extent to which these different explanations are
applicable may vary across taxa (Petrie & Jennions, in press).
While individual repeatability of mate choice is a prerequisite for heritable genetic
variation in preferences (but see Butlin, ), repeatability may also be due to non-
heritable factors. Jennions et al.() reported repeatable phonotactic responses in a
frog (H.marmoratus). However, this did not appear to be due to variation in preference
functions, but rather a size-based ability of females to discriminate between signals.
Similarly, in the frogs U.laevigata and Ololygon rubra there is very strong assortative
mating with respect to body size (Robertson, , ; Bourne, ), suggesting
that repeatability of mate choice for different-sized males is due to variation in female
size. If body size is a consequence of food availability during ontogeny rather than
genetic variation (e.g. Emlen, ), there will be non-heritable, but repeatable, female
preferences for different-sized males. On the other hand, if female body size has a
heritable component, this effectively represents a source of genetic variation in mating
preferences.
In some cases, epigenetic factors mimic genetic heritability, even when breeding
experiments are performed. ten Cate & Bateson (,) have noted that sexual
imprinting in birds may lead to preferences for conspicuous male characters that are
inherited. A preference for novel mates combined with an asymmetry in response may
lead to a preference for ‘ supernormal stimuli ’ (see Bakker, ). For example, a female
sexually imprinting on her father who is long-tailed may preferentially mate with
slightly longer-tailed males. In turn, her daughters, because their father is long-tailed,
will also show a preference for longer-tailed males. A significant argument has also been
proposed in terms of the ‘ peak-shift’ effect, whereby learning to discriminate between
negative and positive stimuli leads to stronger responses to stimuli of a greater
magnitude than the initially preferred stimulus (Weary, Guilford & Weisman, ).
 M. D. J  M. P
In several birds, females tend to remate with the same male either within or across
seasons [e.g. cock-of-the-rock Rupicola rupicola (Trail & Adams, ) ; white-bearded
manakin Manacus manacus (Lill, ); golden-headed manakin Pipra erythrocephala
(Lill, ); Lawes’ parotia Parotia lawesii (Pruett-Jones & Pruett-Jones, ); and
black grouse Tetrao tetrix (Rintama
$ki et al., )]. In general, learning the identity of
males and then remating with a familiar male may also explain some cases of
repeatability of mate choice.
An important methodological issue not addressed in previous studies of repeatability
is that variation in female receptivity or motivation can alter estimates of repeatability
(see Rowland et al., ). They found that female sticklebacks (Gasterosteus aculeatus)
which responded strongly to playbacks of a male displaying (shown in either colour or
greytones) clearly and repeatably preferred a coloured display, whereas those that
responded less showed no consistent preference for coloured over greytoned displays.
It will be interesting to test repeatability over longer time periods to see whether
variation among females is due to short-term effects, or consistent preferences that are
stable over time. Longer-term studies seem preferable if the aim is to estimate genetic
heritability. Non-heritable variables which potentially affect female choice should also
be controlled for. For example, age is known to influence mate-choice decisions (e.g.
insects: Watt, Carter & Donohue,  ; Ritchie,  a; fish: Dugatkin & Godin,
).
When estimates of repeatability are used to calculate phenotypic variation among
females, there is an implicit assumption that female preference functions are being
expressed. However, estimates of repeatability, as with those of heritability, vary
depending on the environment in which they are calculated. For example, females may
express repeatably different preferences in the laboratory, but costly preferences may
not be expressed in a more natural setting; leading to all females in the field mating
randomly with respect to the tested male trait. Conversely, all females may rank males
in the same order in a laboratory experiment. However, the same females may vary in
their mate-choice decisions when sampling or assessment costs are increased (as is likely
in a natural setting), leading to a positive estimate of repeatability. At present, several
research techniques are aimed more at establishing preference functions than total
mating preferences. For example, a common neurobiological approach involves the
description of auditory tuning curves. Aside from concerns that sensitivity need not
indicate preferences (W. Eberhard & W. Wcislo, in preparation), even if tuning curves
are reliable predictors of behavioural preferences, differences in choosiness also affect
mate choice. The willingness of individuals to translate preference functions into mate
choice cannot be directly established using data on peripheral sensory systems
(Dawkins & Guilford, ). Hence, the absence of variation in tuning curves cannot
be equated with the absence of variation in mating preferences. Hopefully, this
illustrates our claim that the distinction between ‘ preference functions’ and
choosiness’ is not just semantic. Ideally, we suggest that mate choice be examined
under laboratory conditions which will reveal ‘idealized’ preference functions, and in
the field (or less benign laboratory conditions) where variation in choosiness may
provide a more realistic estimate of repeatability of mating preferences.
Studies of repeatability are particularly useful when they test for relationships
between female traits (some of which may be heritable) and patterns of mate choice.
Variation in mate choice: a review 
The concept of ‘mating-preference genes’ is highly abstract because so little work has
been conducted into the proximate mechanisms generating mate-choice decisions.
What traits do these genes influence? It is thus difficult to conceptualize mechanisms
that maintain population variability in mate choice, especially if some preferences are
more costly than others. There is, however, growing evidence that mate choice is related
to female body size and body condition (see Section VIII). So familiar, heritable traits
such as body size, parasite resistance, growth rate and time of emergence may all play
an important role in maintaining heritable variation in mating preferences. There is
already a large literature explaining how genetic variation in these traits is maintained
(e.g. frequency-dependent selection, temporal and spatial changes in selection).
Demonstrating a relationship between heritable traits and mate choice may thus
provide an important ‘short cut’ to explaining variation in mating preferences. To date,
most work has been correlational (see Section VIII); experimental manipulations of
body size, body condition, parasite load and other traits to determine their effects on
mate choice are now needed. Of course, we also need to understand why these traits
influence mate choice. The effect they have on reducing the costs of choosiness is an
obvious area for research.
Even when heritable variation in mating preferences is absent, genetic effects may
still be important. The indirect benefits of choosing particular mates often vary in
relation to an individual’s genotype, due to differences in genetic compatibility.
Inbreeding depression (Thornhill, ), optimal outbreeding, kin recognition
(Bateson, ), risk of hybridization and increasing variability at histocompatibility
loci (Potts, Manning & Wakeland, ) have all been shown to affect mate choice. The
influence of these more ‘ extreme’ genetic effects has yielded a slew of data on mating
preferences. Genealogy is often an important aspect of these studies, so the focus is on
individuals and their relatedness rather than population averages. For example,
research into selective mating based on the major histocompatability complex (MHC)
shows that individual females prefer to mate with partners with different MHC types
(reviewed in Zuk,  ; see also Wedekind et al.() for a recent experimental study
in humans). These studies also provide vital data on the ontogeny and proximate basis
of mating preferences. For example, work on kin recognition shows that the likelihood
of mating is partly dependent on the degree of relatedness between potential mates.
This is often a consequence of experiences during infancy such as being reared in the
same litter, rather than an innate ability to detect kin (Fletcher & Michener, ).
Similarly, Simmons () has shown that an ‘ innate’ mating bias against kin in the
cricket Gryllus bimaculatus is strengthened when females experience the pheromones of
unrelated males. If the same attention is paid to the history of individuals in studies of
mate choice for ornaments, equally interesting findings are likely to emerge (e.g.
Barlow, Francis & Baumgartner, ).
V. NON-HERITABLE CAUSES OF VARIATION IN MATE-CHOICE DECISIONS
Random mating is often taken as indicative of an absence of mating preferences
(Pyron, ). However, this is unlikely, as preference functions leading to different
ranking of males probably occur in most species due to inevitable biases built into
sensory systems (Enquist & Arak,  ; Arak & Enquist, ). Whether these
preference functions translate into mate-choice decisions depends crucially on whether
 M. D. J  M. P
there is also selection for female choosiness (Dawkins & Guilford, ). If females
mate with the first male encountered, sensory biases are less likely to generate non-
random mating, unless these biases affect which males are detected first (but see Arak,
). This is a point that has been neglected in some recent discussions of sensory
exploitation. Here, we describe some problems with interpreting mate choice in terms
of mating preferences. Our review illustrates two main points. First, that some factors
lead to mate-choice patterns that mimic heritable variation in mating preferences.
Second, that other factors obscure heritable variation in female preferences. We divide
these factors into three overlapping classes : environmental, social and female
phenotype.
VI. ENVIRONMENTAL FACTORS
()Time and energy costs of sampling
Costs associated with mate sampling are a potential constraint on optimal mate choice
(Gibson & Bachman, ). In some cases, females may therefore utilize cost-reducing
tactics (Ho
$glund & Alatalo, ) which will sometimes reduce the intensity of sexual
selection on male traits. For example, female three-spined stickleback (G.aculeatus)
more readily accept a dull coloured male as a mate when forced to swim against a strong
current between successive males (Milinski & Bakker, ). Slagsvold et al.()
found that female pied flycatchers (F.hypoleuca) move comparatively short distances
when sampling mates, and that the total distance travelled is positively related to
ambient temperature. Females move shorter distances when it is colder. They suggest
that costs associated with sampling may even cause some females to choose an already
mated male (Slagsvold & Dale, ). When the distance between nest boxes is
increased, females show reduced choosiness as fewer males are sampled. Females are
also prepared to suffer direct fitness costs, as reduced sampling leads to more matings
with males with nest boxes near the ground. These nestboxes are more likely to suffer
predation (Alatalo, Carlson & Lundberg,  a). In the peacock wrasse (Symphodus
tinca) whether or not females seek matings with those males that provide parental care
is related to search costs (Warner et al., ). A theoretical model relating the number
of nests sampled prior to mating to the costs of searching and the benefits of parental
care closely fitted the predicted changes in female choice behaviour over the breeding
season. Female choosiness showed phenotypic plasticity in response to temporal
changes in search costs. In general, the spatial distribution of suitable breeding sites has
been analysed in terms of the effect on mating systems with regard to the opportunities
for males to ‘monopolize’ females (Davies, ). If females are sensitive to sampling
costs, however, male density and distribution will also determine other aspects of
mating structure by limiting the scope for active female choice.
Very few studies have attempted to calculate directly the energetic costs associated
with mate choice in the field. Those that have suggested that they are often not that
high. In sage grouse, Centerocercus urophasianus, travelling to leks increases daily
energy expenditure by only % (Gibson & Bachman, ). In other cases, energetic
costs do appear to influence mate choice (e.g. Milinski & Bakker, ). There is greater
evidence that time constraints are important (Sullivan, ,) because : (i) there is
an optimal or limited period in which young can be produced. In temperate birds, for
example, reproductive success declines with the date of initiation of breeding (Møller
Variation in mate choice: a review 
a; see also Backwell & Passmore, in press). (ii) Females are only capable of being
fertilized for a short period of time due to physiological constraints. In anurans, for
example, females that do not obtain a male will oviposit even in the absence of a male
(M. D. Jennions, personal observation). In black grouse, T.tetrix, females show a
definite desire to mate at certain times (see Ho
$glund & Alatalo, ). (iii) Females risk
failing to obtain a mate (see Møller,  a; Palokangas, Alatalo & Korpimaki, ).
In pied flycatchers, F.hypoleuca, the breeding status of males often changes between
successive sampling visits by the same female. Indirect female–female competition for
access to mates may also impose time constraints on sampling (see Section VII.).
Although time constraints exist, females generally benefit by spending not only more
time sampling more males, but also more time assessing each male (e.g. Getty, ).
The accuracy of assessment of mate quality probably increases with the amount of time
spent monitoring behaviour (Sullivan, ,).
()Predation risk
Predation risk has pronounced effects on female choosiness (Magnhagen, ).
Field studies have revealed predation costs associated with mate searching in numerous
species [e.g. cicadas (Cicadetta quadricincta) (Gwynne, ) ; waterstriders (Gerris
remigis) (Sih, Krupa & Travers, ); and isopods (Paracerceis sculpta) (Schuster,
); but see Gibson & Bachman ()]. Theoretical studies predict that the effort
committed to discriminating among mates will decrease as predation risk increases
(Hubbell & Johnson,  ; Crowley et al., ). It has been suggested that variation
in mating preference among populations in guppies (P.reticulata) is partly due to
differing risks of predation (Houde, ). Females in populations with major fish
predators show a reduced preference for males with more orange colouration (Houde
& Endler, ; Endler & Houde, ). One recent study suggests a possible cost to
sampling males. Female guppies near brightly coloured males are more likely to be
eaten by a predator than are the males (Pocklington & Dill, ).
Female crickets (G.integer) were offered a choice between long-bout- and short-
bout-duration calls in an open phonotaxis arena. They all chose the long-bout call
(Hedrick & Dill, ). However, if cover in the form of cardboard was provided on the
side broadcasting the short-bout calls, some females switched and chose the short-bout
call. The proportion of females choosing the short-bout call was positively correlated
with the amount of cover provided. Assuming that staying under cover reduces the risk
of predation, this study suggests that female choosiness decreases as predation risk
increases. Preference functions did not appear to change. Females that chose the short-
bout call when it was in the direction of cover were re-tested immediately afterwards
in an open arena. Most still chose the long-bout call. In an earlier study, Backwell &
Passmore () examined the effect of environmental variation on female phonotaxis
in the reed frog H.marmoratus by placing perches in one half of a test arena. Two
identical calls were broadcast from opposite ends of the arena, and most females
approached the call broadcast from the side containing perches. Moreover, although a
significant proportion of females usually approach calls of higher intensity (Bishop,
Jennions & Passmore, ) or lower frequency (Jennions et al., ), the tendency for
females to use perches was sufficiently strong to eliminate this bias when the ‘less
preferred’ call was broadcast from the side with perches. The situation was further
 M. D. J  M. P
complicated because the bias for using perches was only evident under reasonably high
light levels (full moon) and not under low light conditions (starlight) even though
females still use perches in near total darkness. It is not clear why females preferred
perches when approaching a call, although reduced predation risk compared to travel
on the ground is an obvious explanation. A. S. Rand (personal communication) has also
found that light level alters female choice in the tungara frog, Physalaemus pustulosus.
Light level is implicated in the perceived risk of predation because males reduce calling
under low-light intensity conditions when it is more difficult to detect predatory bats
(Ryan, ).
In a pipefish (Syngnathus typhle), mate choice was measured in a two-choice
aquarium experiment (Berglund, ). In this sex-role reversed species, males choose
females. In the absence of a predator, males courted and mated a large female
significantly more often than a small female. When a predator was visible, however,
males courted both females equally and mated slightly more often with the smaller
female. Berglund () interprets this as a consequence of males being ‘ less choosy ’.
What exactly does ‘choosy’ mean? Males were not less choosy in terms of a reduced
willingness to sample mates. Both females were continuously visible, and males were
actually more active when a predator was present. When both potential mates are
instantly’ available, random-mating suggests that the males no longer have a
preference function which ranks a large female above a small female. This may be due
to a greater risk of predation when associated with a larger female who is probably a
more profitable prey item. Another explanation for the apparent change in preference
function is that males were more willing to mate because their estimate of residual
reproductive value was lowered. They may have moved between females more often in
the hope that at least one of them would deposit eggs. It would be interesting to know
whether there is variation among males in the effect of predator presence. Mating was
random’ in the sense that approximately equal numbers of males approached large and
small females. However, this could either be because all males ranked large and small
females equally, or because a proportion of males now ranked smaller females above
larger ones. A very similar effect of predator presence on mate choice was also shown
for female choice of males in the sand goby (Pomatoschistus minutus) (Forsgren, ).
Field studies also suggest that predation risk can influence mate choice. In Uganda
kob (Kobus kob thomasi), females preferentially mate on leks with good visibility and
within leks with males holding territories with better visibility (Deutsch & Weeks,
). Under these conditions, mating preferences based on male phenotype may not
be expressed because of direct benefits to females mating in areas where predation-risk
is lowered.
()Territory or resource quality
Mate choice is often based on both male phenotype and the quality of the territory
or resources defended by males [topi (Damaliscus lunatus), puku (Kobus vardoni)
(Balmford, Rosser & Albon, ) ; pied flycatcher (F.hypoleuca) (Alatalo, Lundberg
& Glynn,  ; Lifjeld & Slagsvold, ); two species of fish, Forsterygion varium
(Thompson, ) and Ophioblennius atlanticus (Co
#te & Hunte, ) and a fiddler
crab, Uca annulipes (Backwell & Passmore, in press)]. Predation risk may also vary
between territories and may have a greater influence on mate choice decisions than
Variation in mate choice: a review 
variation in male phenotype (Deutsch & Weeks, ). Whether or not mating
preferences for male phenotypes are expressed depends on the distribution of material
resources in relation to male phenotype. For example, when material resources are
given priority, variability in their distribution is likely to obscure preferences for male
phenotype. Lifjeld & Slagsvold () found that the pattern of female mate choice in
pied flycatchers (F.hypoleuca) was non-random with regard to male plumage when the
habitat was homogeneous. In contrast, earlier work in a heterogeneous environment
showed that female choice was unrelated to male phenotype (Alatalo et al., ).
Changes in female choosiness in relation to different preference functions will also
influence mate-choice patterns. For example, females can continue sampling until they
mate with a male with a highly ranked phenotype on a highly ranked territory, or they
can reduce their mating threshold for one or more of the mate-choice cues. Variation
in the costs of choosiness may therefore lead to spatial or temporal variation in the
strength of the association between male phenotype and mating success. For example,
in the fiddler crab U.annulipes, early during each  day semi-lunar cycle females
mated with larger males whose burrows exceeded a threshold value. Females did not
return to previous burrows and the mated male’s burrow invariably had a higher
quality index’ than any of the other burrows sampled. Later in the breeding cycle there
was no large-male mating advantage, but females still chose to mate in burrows with the
highest ‘quality index’ (Backwell & Passmore, in press). Thus, there appears to have
been a change in female choosiness for male size but not for burrow features.
()Environmental effects on signal detection and discrimination
Earlier we defined preference functions as the order in which phenotypes were
ranked. Of course, the shape of the function also influences mate choice. The potential
for reversals in the relative ranking of phenotypes increases when functions have flat
gradients and there is imperfect assessment of signals (Johnstone,  ; Wiley, ).
These errors in assessment may lead to changes in mate-choice decisions. For example,
another explanation for the results of Berglund () (see Section VI.) is that the
equal mating success of large and small female pipefish with a predator present was due
to males focusing their attention on monitoring the predator rather than assessing the
females. Variation in ‘attention’ affects discriminatory ability in foraging and learning
situations and is likely to have equally important effects in mate choice. Getty ()
provides a recent and interesting review of the optimal period that females should invest
in discrimination. His model suggests that when different contexts lead to different
levels of selectivity, individuals should assess their own ROC (Receiver Operating
Characteristic) curves (which indicate discriminatability) in relation to ‘ their own
sensory, perceptual and cognitive constraints’.
Sensory capabilities are influenced by environmental conditions which determine the
accuracy of signal detection and discrimination. For example, it is impossible to
discriminate between certain colours under certain light conditions (e.g. Milinski &
Bakker, ). Over evolutionary time, there has been selection on males to use those
channels of communication most suited to the environment (Marchetti, ). Acoustic
signals are particularly sensitive to attenuation and degradation which varies
systematically between different habitats. This has led to consistent differences in call
structure associated with different habitats (Wiley, ). Similarly, when the ability to
 M. D. J  M. P
discriminate signals varies among habits, females may give preferential weighting to
those signals that are most easily discriminated. This may lead to spatial and temporal
variation in the characters which influence mate choice.
In several frog species, mating preferences observed in the laboratory are not
reflected in the pattern of mating in the field (Gerhardt, ). For example, laboratory
phonotaxis trials indicate a strong female preference for low-frequency calls in reed
frogs (H.marmoratus). Although larger males produce lower-frequency calls, a field
study of male mating success involving over  males failed to demonstrate a large-
male mating advantage (Dyson et al., ). In laboratory studies, the ability of female
frogs to discriminate calls differing in intensity (Bishop et al., ) or frequency
(Gerhardt, ,) is reduced if the number of speakers broadcasting calls is
increased. This suggests that the inability of some frogs to express mating preferences
seen in the laboratory may be attributable to the increased acoustic and structural
complexity of the natural setting (Telford, Dyson & Passmore,  ; Gerhardt, ).
Evidence that female choice can explain non-random mating is sometimes based on
mate choice under artificially simple conditions predicting the observed field mating
pattern (e.g. a large-male mating advantage). This conclusion may well be correct, but
it should be supported by evidence of mate choice in the field. In the case of phonotaxis
studies on frogs and insects, it is relatively easy to repeat key laboratory experiments by
broadcasting calls in the field and counting the number of females that different stimuli
attract. Surprisingly, this rather obvious approach has rarely been used (but see
Gibson, ).
Gerhardt & Klump () showed that green treefrogs (H.cinerea) can only detect
calls with an intensity equal to or greater than that of the background chorus (see also
Schwartz & Gerhardt, ). In a large breeding chorus, this probably limits females
to detecting three to five males at any given moment. Unless females actively move
around the chorus, mate choice may thus be based on a far smaller set of males than
those present at a breeding site. Knowing which males a female is capable of detecting
has important implications when determining mating preferences (e.g. Morris, ).
Forrest & Raspet () provide a detailed model illustrating the difference between
passive and active choice and discuss female choice in relation to male spacing in
acoustic signallers. They note that the results of phonotaxis studies are difficult to
interpret in terms of female sampling tactics without knowledge of female sensory
capabilities. In general, better understanding of female neurobiology and sensory
capabilities is a prerequisite for any real advance in our understanding of mate choice
(see Bennett, Cuthill & Norris, ).
How can we distinguish perceptual ‘errors’ from variation in preferences? Two lines
of evidence are available. First, repeatability of female choice should be low if females
cannot distinguish stimuli, and mate choice is therefore random with respect to actual
mating preferences. Second, if there is a strong bias at the population level for certain
signal features, but this preference disappears when the variability in these signals is
reduced, then lack of discrimination (rather than lack of preference) is probably
responsible (although lack of discrimination is not the only possible explanation for
such an effect because reduced variability also lowers the benefits to be gained from
choosing). However, even when there is no population-level bias, some females may
still be discriminating. Females may vary in their general ability to discriminate
Variation in mate choice: a review 
(Jennions et al., ), or in their willingness to invest time in discriminating. If
discriminating females are rare, there may be no detectable population-level mating
bias when sample sizes are small. However, tests of repeatability of choice may still
allow researchers to detect the presence of these individuals. In our experience, studies
are usually constrained by limited numbers of receptive females rather than by the time
available for testing. Repeated testing of individual females may therefore be a
worthwhile activity while waiting for additional receptive females.
VII. SOCIAL FACTORS: THE EFFECTS OF CONSPECIFICS
()Interactions between males
Male–male competition obviously affects female choice if it physically prevents
females from mating with preferred males (Trail, ). However, more subtle
interactions between males can also obscure or alter female mating preferences. One
well-studied phenomenon in acoustics is a ‘precedence effect’ whereby females
preferentially approach the leading call when two calls are presented in a leader–follower
sequence (Greenfield,  ; Minckley & Greenfield, ). This effect may override
other preferences. In H.marmoratus, when calls are presented in a leader–follower
sequence, females prefer the leading call even if it has a higher frequency (Dyson &
Passmore, a; Jennions, ). When calls are presented antiphonally, females
preferentially approach the lower-frequency call (Jennions et al., ). Leader–
follower sequences are generated through male–male interactions, so changes in male
behaviour induced by other males affect female mating preferences for low frequencies
(Greenfield, ). Females also influence males, although the precise adjustments in
male call timing vary among species in a manner which appears to depend on the exact
nature of the female preference for leading calls (see Minckley & Greenfield, ). In
general, in acoustically signalling species there is considerable potential for interference
between callers. Consequently, there are often very precise interactions between males
which act to preserve the structure of their signals (Schwartz, ), and thereby
influence the subsequent ability of females to discriminate between stimuli (Schwartz,
).
Inter-male spacing also influences mate choice in several acoustically signalling
species (e.g. H.marmoratus: Telford, ;Bufo calamita: Arak, ;Tettigonia
viridissima: Arak, Erikkson & Radesater, ). There may be trade-offs between
female preferences for call features and for males that are either clumped or dispersed.
Different forms of male–male spacing and chorus organization therefore have strong
effects on patterns of mate choice for call features. It is difficult to predict how male
spacing will influence mate choice because this depends on both the choice tactics
females use as well as on their sensory capabilities (reviewed by Forrest & Raspet,
). Female sensory capabilities determine both the number of males that are audible
(Gerhardt & Klump, ) and the extent to which females can discriminate between
neighbours (Telford, ).
In the field, male reed frogs (H.marmoratus) do not generally interact acoustically
with one another once they have started advertisement calling. There is no consistent
call alternation or synchronization (Dyson, Henzi & Passmore, ). Consequently,
leader–follower patterns emerge and switches in leadership between males often occur
by chance. Females prefer leading calls as well as low-frequency calls. Given these two
 M. D. J  M. P
preferences how do they locate males ? Continual approaches towards a leading call
would be time-consuming if the location of the leader continually changes. In two-
choice phonotaxis, females were allowed to begin their approach to a leading call. The
leading and following calls were then switched between speakers (Dyson et al., ).
The likelihood that the female reorientated and continued to approach the leader was
distance-dependent. At greater distances, they were less likely to reorientate. In
addition, the frequency of the calls also had an effect. When a high-frequency leader
was switched to a follower and the low-frequency follower to a leader most females
reorientated and approached the low-frequency leader. In contrast, they were less likely
to reorientate when a higher-frequency follower was switched to a leader and often
continued to approach the now low-frequency follower call. This work summarizes
three important points. First, male spacing determines the likelihood that females
reorientate towards a new male. Second, male–male interaction patterns determine
whether or not the female preference for leader}follower or high}low frequencies is
expressed. Third, interactions between preferred traits are often synergistic (see
Section IX.).
Perhaps the most dramatic effects of male spacing patterns on female choice arise
with leks (Ho
$glund & Alatalo, ). The causes of lekking probably vary from species
to species. In ungulates, for example, Clutton-Brock, Deutsch & Nefdt () suggest
that the tendency for females to mate on leks is a response to the high risk of harassment
from other males when matings are performed in large herds. Other workers have
suggested that leks arise because females prefer to mate in places where they can more
easily assess several males (Bradbury, ). What is evident, however, is that mate
choice on leks is often based on phenotypic and environmental features that differ from
those used away from leks (Balmford, ). One possible reason for this is that
sampling on leks is less costly, allowing females to use more discriminatory sampling
tactics (Janetos, ) such as ‘ best-of-N’ sampling which has been recorded in most
lekking species (see Section IX.). In contrast, this tactic is probably infeasible in, say,
a large herd of ungulates and is less likely, or at least the number of males sampled will
be far fewer, when males are widely dispersed. On leks, copying the mate choice of
others is also more feasible than in other mating systems (Gibson & Ho
$glund, ).
Most of the species in which mate-copying has been identified are lek breeders (see
Section VII.).
While leks are usually associated with greater scope for female choice, recent
blackhole’ models suggest that female ungulates may mate indiscriminately on leks
(Clutton-Brock, Price & MacColl, ; Stillman, Clutton-Brock & Sutherland, ).
In these models, lekking and non-random patterns of mating are determined by the
relative ability of males to sequester females. Because males on leks are closer together,
not only active female choice but also direct male–male competition are more likely. On
larger leks, for example, there is a reduction in male mating skew that may be related
to increased levels of male aggression (Widemo & Owens, ). Whether greater
male–male competition hinders or benefits females depends on the extent to which male
dominance correlates with female mating preferences. Several studies of lekking species
suggest that females do prefer dominant males (Gibson & Bradbury,  ; Alatalo,
Ho
$glund & Lundberg,  ; reviewed by Ho
$glund & Alatalo, ).
Variation in mate choice: a review 
()Variability in male phenotypes
Females can only express mating preferences if there is sufficient phenotypic
variation among males to detect differences (Petrie, ; Cherry, ). Interestingly,
sexual traits often show greater phenotypic variation than ordinary morphological traits
(Alatalo et al., b; but see Barnard, ). Evidence that variation is important
comes from field studies where mating is naturally random or weakly skewed with
respect to ornament size, but where experimental manipulation of ornaments that
increases size variation results in non-random mating (e.g. longtailed widowbirds,
Euplectes progne, Andersson,  ; reviewed by Ryan, ). Sullivan () suggested
that female choice is usually based on the most variable male signals because these traits
show the strongest correlation with male mating success (e.g. Ipswich sparrows,
Passerculus sandwichensis: Reid & Weatherhead, ; Great snipe, Gallinago media:
Fiske, Ka
/la
/s & Sæther, ) E. Forsgren (in preparation) found that female sand
gobies, Pomatoschistus minutus, are more choosy when variability in male courtship
levels is higher. However, it is more difficult to obtain a statistically significant
relationship between mating success and male phenotypic traits when variability in the
trait is low (N. Wilson, personal communication), and this may also account for the
observed pattern. In a model, Real () showed that choosiness increases when there
is greater variation in male quality because of the potential increase in benefits
associated with mating with a high-quality partner (see also Getty, ). If male
quality is signalled phenotypically, then greater phenotypic variability is also associated
with greater potential benefits of choosiness.
()Female–female competition
The role of male–male competition and interference in reducing a female’s ability to
express mating preferences is well known (Thornhill,  ; but see Trail, ).
However, with the exception of potentially sex-role reversed species such as doterels
(Charadrius morinellus) (Owens, Burke & Thompson, ), far less attention has been
paid to female–female competition. Female–female competition is not only associated
with sex role reversal, and may occur in any situation in which there is large variation
in male quality (e.g. Petrie, ). In lek-breeding peacocks (Pavo cristatus), dominant
females attempt to monopolize preferred males by repeatedly engaging them in
courtship. This results in some subordinate females mating with males with smaller
trains (Petrie et al., ). In colonially breeding razorbills (Alca torda), extra-pair
copulation (EPC) occurs on lek-like mating arenas. Females sometimes disrupt EPC
attempts by their mates using direct physical aggression towards both their own mate
and the extra-pair female (Wagner, ). Most mating-system studies focus on the
role of females in mate choice, but direct female–female aggression is also important
and seems to occur in many species (see Ahnesjo
$et al., ; Rosenqvist & Berglund,
). In birds, it may be an important factor in the maintenance of monogamy
(reviewed by Slagsvold & Lifjeld, ; Eens & Pinxten, in press).
Female competition may sometimes take subtle and indirect forms that do not
involve overt aggression. In pair-bonded birds, females may engage in repeated
copulation with their mates to reduce the risk of male EPC or of a second female settling
on their territory (Petrie, ; Petrie & Hunter, ; Hunter et al., ). When
 M. D. J  M. P
mating is positively assortative with respect to quality, and higher-quality females are
better at preventing EPC, this may make it more difficult for low-quality females to
mate with high-quality males. There is evidence both for assortative mating on the basis
of ‘quality’ and that females preferentially seek EPC with higher-quality males (e.g.
Houtman, ; Kempenaers et al., ).
There have, however, been few attempts to test whether females vary in their ability
to prevent EPC or the formation of additional pair-bonds by their partner. What traits
or tactics might be responsible for such variation (Petrie & Hunter, ; Slagsvold &
Lifjeld, ; Kempenaers, )? Whittingham, Dunn & Robertson ()
investigated the ‘multiple copulation as mate guarding’ hypothesis in tree swallows
(Tachycineta bicolor) and concluded that it did not explain why females mate multiply
(but see Eens, Pinxten & Kempenaers,  ; Whittingham, Dunn & Robertson, ).
In contrast, there is good evidence from both observational (Eens & Pinxten, ) and
experimental studies (Eens & Pinxten, in press) that female European starlings (Sturnus
vulgaris) do mate guard by soliciting copulations. When a second female was introduced
to a caged pair, the original female increased her rate of copulation solicitation. It
remains to be seen, however, whether females vary in the success with which they
employ this and other tactics.
In some poison-dart dendrobatid frogs (e.g. Dendrobates auratus), female–female
competition is not associated with classic sex-role reversal. Rather, it seems to involve
a defending female preventing other females from mating with a male with whom she
has previously mated (Summers, ,). This may occur because per capita
tadpole survival is lowered when males tend more than one brood (Summers, ).
Repeated female courtship of males in dendrobatid frogs has also been interpreted as
an attempt to prevent males from mating with additional females (Summers, ).
Female brentid weevils (Brentus anchorago) also disrupt courtship and copulation
attempts. This ‘spiteful’ behaviour is thought to reduce the number of females
ovipositing and increase larval survival rates (Johnson, ). In flycatchers (F.
hypoleuca), resident females prevent or delay other females from settling which
increases the amount of paternal care their own offspring receive (Slagsvold & Dale,
; Dale & Slagsvold, ). Female–female aggression may also limit the ability of
females to search for mates (Dale, Rinden & Slagsvold, ). Similar female–female
aggression has also been reported for blue tits, Parus caeruleus (Kempenaers, ,
). The general effect of these female–female interactions in dart-frogs, weevils and
birds is potentially to constrain the expression of mating preferences of at least some
females (see also Møller,  a). The outcome of these constraints in generating
variation in female fitness is not well studied. However, when choice is for direct
benefits, the costs of a sub-optimal choice are likely to be considerable. Even in species
in which there are no obvious material benefits associated with mate choice (but see
Fox, McLennan & Mousseau, ), choice may have benefits in terms of increased
offspring viability (Norris, ; Petrie, ). The inability to choose freely may
therefore generate substantial variation in offspring fitness.
()Female mate copying
Recent studies indicate that individual mate-choice decisions are not always
independent and that females may copy one another (reviewed by Pruett-Jones,  ;
Variation in mate choice: a review 
Gibson & Ho
$glund, ;Ho
$
glund & Alatalo, ). Earlier theoretical work on
lekking species suggested mate-choice copying because unanimity of choice appeared to
be greater than expected on the basis of independent female choice (Bradbury,
Verhencamp & Gibson, ). Mate copying has been implicated in skewed mating
patterns in two species of manakin (Pipra erythrocephalus,Manacus manacus) (Lill,
,), black grouse (Tetrao tetrix) (Ho
$glund et al., ), sage grouse
(Centrocercus urophasianus) (Gibson, Bradbury & Verhencamp, ), Lawes parotia
(Parotia lawesii) (Pruett-Jones & Pruett-Jones, ), fallow deer (Dama dama)
(Clutton-Brock et al., ; Clutton-Brock, Hiraiwa-Hasegawa & Robertson, ;
but see McComb & Clutton-Brock, ), guppies (P.reticulata) (Dugatkin, ) and
mollies (Schlupp, Marler & Ryan, ). There is also evidence from two species of
antelope (Kobus kob,K.leche kafuensis) that females preferentially mate on lek
territories where other females have previously been. Experimental relocation of soil
suggested that females use substances contained in urine to assess earlier female
presence, and preferentially mated on territories with higher female visitation rates
(Deutsch & Nefdt, ). Most cases in which mate copying is thought to occur involve
lekking species.
Mate-choice copying has been explained in at least two ways. First, as a tactic that
reduces the costs of sampling. Although sampling on leks does not appear to be costly
in terms of energetics (e.g. Gibson & Bachman, ) or predation risk (see Ho
$glund
& Alatalo, ), time constraints may be important (see Section VI. ). Losey et al.
() developed a simulation model in which both the costs and benefits of copying
differed from those of direct mate assessment. They showed that due to frequency-
dependent selection (not everyone can copy), copying and direct assessment may yield
the same overall fitness payoffs. If copying does reduce costs, we might expect females
to be more likely to copy when they are vulnerable to sampling costs. Thus, females in
poor condition, with high parasite loads or that are inexperienced should be more likely
to copy (Pruett-Jones, ). Dugatkin & Godin () found that young guppies (P.
reticulata) were more likely to copy the mate-choice decisions of older guppies. In sage
grouse (C.urophasianus) and black grouse (T.tetrix), younger females mate later than
older females and are thus more likely to be the copiers (Ho
$glund & Alatalo, ).
Age-related changes in propensity to copy may have important implications.
Kirkpatrick & Dugatkin () modelled the copying of older females by younger ones
where preferences were ‘ culturally’ acquired. The outcome was the coevolution of
preference and trait and a decreased likelihood that female preferences will maintain
novel male traits. In general, if female preferences become more directional with age or
experience due to increased investment in sampling (i.e. greater choosiness) or through
developmental changes in preference functions then copying may strengthen the
intensity of directional sexual selection on preferred male traits. Experimentally, there
seems to be room for future work manipulating the costs of mate choice to investigate
whether copying can be increased. Parasite load, body condition and perceived parasite
risk could all be altered in laboratory experiments with, for example, guppies to
determine the effect on copying behaviour.
Second, copying may be a mechanism whereby females can quickly obtain more
precise information about potential mates. If the mate-choice decisions of others
contain information about specific males, then watching the mate choice of others
 M. D. J  M. P
decreases the likelihood of choosing an inappropriate mate (Bikchandani, Hirschleifer
& Welch, ). However, this may also lead to so-called ‘ information cascades
(Gibson & Bachman, ) in which an increase in reproductive skew may arise due to
amplification of initially small variations in female visitation rates that are unrelated to
preferred male traits (Deutsch & Nefdt, ). Whether copying leads to stronger
directional selection or arbitrary trends depends crucially on whether the mate choice
of copied females is random or not. It is less likely to be random when some females
copy and others actively choose. ‘Information cascades’ are more likely when all
females show a propensity to copy (Bikchandani et al., ). Again, we need data on
variability among females. How much variability is there among females in their
propensity to copy?
A distinction is often drawn between active copying of mating decisions of other
females and a general preference for association with other females in aggregations
(‘conspecific cueing’) (McComb & Clutton-Brock, ). However, this distinction is
reminiscent of that between active and passive mate choice which was strongly
criticized on the grounds that it confused proximate and ultimate questions (see
Sullivan, ). Both aggregation and copying have the same ultimate effect whereby
variation in male mating success is increased due to females approaching sites where
other females either have been or are present. The proximate mechanisms responsible
may, of course, differ among species and knowing what they are is therefore necessary
for a full understanding of why females influence each others’ mating decisions.
()Density and the operational sex ratio
The operational sex ratio (OSR) and the spatial distribution of the two sexes have
long been key elements in theoretical discussions of sexual selection and the evolution
of mating systems (reviewed by Davies,  ; Clutton-Brock & Parker, ). Even so,
much remains to be discovered about the proximate mechanisms leading to behavioural
changes, and the validity of some predictions of OSR theory are now also questionable
(see Arnold & Duvall, ). For example, inter-sexual competition can occur in both
the choosy and the non-choosy sex (e.g. Petrie et al., ). When females are the more
aggressive sex, this need not mean that female mate choice does not occur, or that male
mate choice will occur (Summers, ; Owens et al., ). The extent of variation
among individuals is an element that should be incorporated into theoretical models
more often. It determines the benefits of mate choice and may also affect the costs of
choosiness. In the same way that the OSR and density influence alternative mating
tactics of males (Lucas & Howard, ), they may also influence alternative sampling
tactics and mate-choice decisions of females.
If the OSR can be used to predict changes in mating behaviour then animals are
either able to estimate the OSR or, more plausibly, there are proximate cues related to
the OSR. How are cues associated with the OSR and density likely to affect mate
choice? One possibility is that they affect the costs of sampling. () When the density
of the chosen sex is lower, there are increased distance, energetic and time costs to
sampling (Real, ). These should lead to a reduction in choosiness. () There is an
increased risk of failure to mate or, in pair-bonded species, of not receiving assistance
with parental care, when the OSR is less biased towards the chosen sex (Møller,  a).
Females may therefore adjust preferences or choosiness in response to the rate at which
Variation in mate choice: a review 
they encounter other females. In pied flycatchers, F.hypoleuca, sampling females may
use the rate at which they encounter other sampling females (Dale & Slagsvold, ).
Females may also become less choosy or change their preferences as they encounter a
greater proportion of males that are already paired. In some species, the OSR may be
such that ‘sex-role reversal’ occurs and the gender of the choosy sex changes (Gwynne,
). Role reversal, however, is only a highlight in a continuum of changes in
choosiness. Even before role reversal, there may be changes in the extent to which less
attractive individuals can express their mating preferences due to increased likelihood
of rejection. The perceived OSR may therefore have effects on sampling and mate
choice that differ depending on a female’s phenotype. Less attractive females may
become less choosy at lower female-biased OSR values than more attractive females
because the costs of choosiness are negatively correlated with attractiveness (see Section
VIII).
An increase in choosiness at higher densities of the chosen sex has been shown in pied
flycatchers (F.hypoleuca) (Alatolo et al., a), a katydid (Kawanaphila nartee) (Shelly
& Bailey, ) and kestrels (Falco tinnunculus) (Palokangas et al., ). Milinski &
Bakker () have also noted that their experimental manipulation of the costs of
sampling in stickleback (G.aculeatus) mimicked variation in male density. However,
higher male density is not always associated with increased female choosiness. At high
male densities or strongly male-biased OSR values, females may be so persistently
courted by males that it is less costly to accept matings than to try to evade males (e.g.
waterstriders, Gerris spp.: Rowe et al., ), leading to a reduction in the intensity of
sexual selection on male phenotypic traits. Allen & Bailey () also found that the
propensity of male crickets (Requena verticalis) to mate did not increase when the
encounter rate with females was experimentally lowered (cf. results of Shelly & Bailey,
).
Several experimental studies on a range of taxa show that the OSR influences
choosiness. In a two-choice test, male pipefish (S.typhle) preferentially mated with a
large female when the OSR was female-biased, but were equally likely to mate with
large or small females when the OSR was male-biased (Berglund, ). In milkweed
beetles (Tetraopes tetraophthalamus), males were more choosy when the OSR was
female-biased, although they continue to engage in contests with other males
(Lawrence, ). In field crickets (Gryllus pennsylvanicus), females were more choosy
when the OSR was male-biased (Souroukis & Murray, ). (For additional examples
see Gwynne,  and Vincent et al., .) More recent studies have also attempted
to explain variation in the OSR in terms of environmental features such as food
availability (Gwynne & Simmons, ), ambient temperature (Ahnesjo
$,) and
parasite levels (Simmons, ). Knowledge of these features may also be used to
predict variation in mate-choice patterns (see Gwynne & Brown, , for an example
of inter-specific variation in response to the same treatment).
VIII. FEMALE PHENOTYPES
Several studies show a relationship between female phenotype and that of their
mates. This may occur because the direct costs and benefits of mating with certain
males differ among females. For example, larger females may require larger males to
ensure that all their eggs are successfully fertilized, or smaller females may be unable
 M. D. J  M. P
to withstand the cost of mating with a large male (Ryan, ; Robertson, ;
Bourne, ). Mechanical constraints on pairing may also lead to positive assortative
mating in some species (review: Brown, ). There is also evidence that short-term
female condition influences mating behaviour. In water mites, Neumania papillator,
female hunger level determines their responsiveness to males. Males exploit the
female’s attractiveness to stimuli associated with food by mimicking cues produced by
prey items (Proctor, ). Hungry females are more likely to mate with a range of male
phenotypes. In bullies (G.breviceps), parasitized females make fewer visits to potential
mates and are more likely to mate with small males than are unparasitized females
(Poulin, ). In a bush cricket (R.verticalis), parasitized females attempt to mate
more frequently, presumably because they wish to obtain nutrients from male
spermatophores (Simmons, ). This leads to increased choosiness by males, which
should, in turn, have a negative effect on the ability of less attractive females to mate
with preferred males.
A female’s phenotype may predict mate-choice behaviour because it affects sampling
costs. In pied flycatchers (F.hypoleuca), there is a positive correlation between an index
of body condition and the distance females travel prior to choosing a mate (Slagsvold
et al., ). In redlipped blennies (O.atlanticus), larger females travelled further to
reach mates, had more scars (which are sustained during sampling by attacks from
damselfish) and mated with larger males. This suggests that they were prepared to pay
higher costs to mate with larger males who provide better parental care (Reynolds &
Co
#te, ). Choudhurry & Black () found that larger, heavier barnacle geese
(Branta leucopsis) females formed more ‘trial liaisons’ prior to pairing; and Rintama
$ki
et al.() found that large female black grouse (T.tetrix) with a high body mass
visited more males prior to mating. In some species, males seek out females for mating
purposes. In the common shrew (Sorex araneus) there are two types of males. Type B
males who are larger at the start of the breeding season are more likely to move and
encounter females, while Type A are more stationary (Stockley,  a). In contrast,
Fiske & Ka
/la
/s() found a trend for larger female great snipe (Gallinago media)to
spend less time on the lek prior to mating than did smaller females. In general, however,
it appears that females with greater energy reserves or larger body size increase the time
period or area over which they sample, leading to variation in mate choice which may
be unrelated to preference functions. There is a need for further experimental
manipulation of female condition, body size, perceived estimates of residual
reproductive value and other traits which should affect sampling behaviour and mate
choice. It is important to note, however, that these effects are unlikely to be apparent
in low-cost, two-choice experimental set-ups in which mates are simultaneously and
instantly available. Researchers should therefore attempt to design experiments in
which sampling costs are more similar to those in the field (e.g. Milinski & Bakker,
).
In species with mutual mate choice, females vary in their attractiveness to males
(Brown, ; Jones & Hunter, ). This should lead to positive assortative mating
with respect to attractiveness (Burley, ). As preferred males drop out of the pool
of potential mates, some females may be forced to mate with less-preferred males
(Brown, ). Less-attractive females may partly solve this problem by seeking EPC
with more attractive males (Møller, , a); Houtman, ; Kempenaers et al.,
Variation in mate choice: a review 
). In contrast, attractive females may become more choosy in their initial choice of
mate, because they stand a smaller risk of not obtaining a high-quality male, or of failing
to mate (Petrie & Hunter, ). If this is true, females must first assess their own
attractiveness relative to that of other females. We are unaware of data collected to test
whether this occurs.
Alternatively, the preference functions of females may differ. If unattractive females
know’ they are unlikely to retain a high-quality mate, they may prefer lower-quality
males over higher-quality males. Mating with a high-quality male carries several
potential costs: a greater risk of desertion; of a secondary female attempting to pair with
your mate and thereby reducing male care for your own offspring ; of disease transfer
from a mate who performs more EPC (Birkhead & Møller, ); of reduced parental
care by attractive males (Burley, ; de Lope & Møller, ; Møller, c).
Variation in a female’s ability to compensate for these costs may lead some females to
choose low-quality males. This may explain repeatable female choice based on tail
length in barn swallows (H.rustica) (Møller,  b). Tail length appears to be an
honest signal of male quality in this species (Møller & de Lope, ). In the absence
of differing costs, it is perplexing why all females do not prefer long-tailed males.
IX. HOW DO FEMALES CHOOSE MALES?
()Sampling tactics
Until recently, the average sexual selection field study consisted of monitoring males
and counting the number of females they attracted. But how do females find these
males? How do they succeed in mating with males that are larger than average, or have
longer tails or louder calls? For these and related questions, there have been few
empirical answers, but several theoretical investigations have been carried out. A major
issue motivating this modelling has been the effect of preference costs on the outcome
of different models for the sexual selection of male ornaments (Pomiankowski, ;
Reynolds & Gross,  ; Iwasa et al., ; Kirkpatrick & Ryan, ; Andersson,
). In many models, the direct cost of choice is assumed to be outweighed by
indirect benefits (e.g. Iwasa et al., ). Whether this is generally the case remains to
be shown. In rhesus macaques (Macaca mulatta), females that spent time in proximity
to lower-ranking males (which is a correlate of copulation rate) suffered increased rates
of attack by dominant males (Manson, ). This suggests that females are prepared
to incur direct costs from mating with lower-ranking males. However, as in most cases
it is unclear whether the benefits are indirect, direct or both. In general, the
applicability of many models of female-preference evolution are sensitive to the costs
and extent of direct selection on choosiness (e.g. see Pomiankowski,  ; Kirkpatrick
& Barton, ). Empirical data may therefore have a major impact on the plausibility
of some mathematical models.
Theoretical studies have also related adjustment of searching behaviour and choice
tactics to variation in predation risk (Hubbell & Johnson, ), mate density (Crowley
et al., ), costs of memory allocation (Hutchinson, McNamara & Cuthill,  ;
Roitberg, Reid & Li, ) and time constraints (Real, ). Theoreticians have
proposed at least six tactics that can be used when choosing mates (Parker, ,,
; Janetos, ; Wittenberger, ; Real, ; Dombrovsky & Perrin, ).
Examples of species in which these different tactics are thought to occur are presented
 M. D. J  M. P
in Table .() Random mating tactic : accept the first mate encountered. () Fixed
threshold tactic: sample sequentially and accept the first mate that exceeds a set
criterion. Mating always occurs with the last male sampled. () Sequential comparison
tactic (Wittenberger, ) : sequentially compare mates until the most recently
encountered is of lower quality than the previous encountered, then accept the
previously encountered male. This always leads to mating with the penultimate male in
a sampling sequence. () One-step decision tactic (Janetos, ) or sequential search
rule tactic (Real, ) : sample until the value of the mate encountered is greater than
that expected from continued searching. The sequential search rule is a refinement of
the one-step decision tactic in that it also considers the costs associated with continued
searching. As with the fixed threshold model, females generally mate with the last male
sampled. The difference is that this male may not have the highest value for the
preferred criteria. The model can be considered as a modification of the threshold
model, in which the threshold varies through time in relation to the costs of sampling
and the expectation of finding a male that will exceed the present threshold. If females
can recall the positions of previous males, and their threshold is subsequently lowered,
they may sometimes return to previously sampled males (Fiske & Ka
/la
/s, ). ()
Pooled comparison (‘Best-of-N’) (Janetos, ): sample Nmales and then accept the
male with the highest value for the preferred trait(s). This is potentially the most
rewarding, but also the most costly tactic (Real, ). () Optimal stopping rule
(Dombrovsky & Perrin, ) : this rule is similar to tactic , but differs technically in
that it makes the more realistic assumption that the choosy sex does not know a priori
the distribution of quality in the chosen sex. The model does, however, assume that a
female knows a priori the total number of samples she can make (this assumption is
shared with tactic ). This model makes several predictions concerning the length of the
sampling period, the existence of a ‘ previous-male effect’ and that this latter
phenomenon will not only be confined to the previous male sampled, but to males
earlier in the sequence as well. The rule is designed to maximize a female’s chances of
mating with the best male encountered but it ignores any costs of choice and treats the
second-best male as no better than the worst and is therefore unlikely to be relevant to
biological male choice.
In practice, it has proved difficult to distinguish which tactics are being used by
females in the field. This may be partly due to variation among females. Tactics need
not be invariant within a species. They vary in their costliness, and not all females may
be able to use the most expensive tactics. Resolving this problem may require data on
successive sampling-mating sequences by individual females, with suitable statistical
controls for age and size-effects. For example, Fiske & Ka
/la
/s() have shown that
experienced great snipe (G.media) females are more likely to return to mate with a
previously sampled male than are inexperienced females. In spite of these difficulties,
several trends have emerged from field and laboratory studies of sampling behaviour.
() Mate choice in many species that sample sequentially involves adaptive searching.
That is, females adjust their threshold for acceptance in relation to the phenotype of
previously sampled males. Choice is relative, not absolute. A ‘ previous-male effect’ has
been reported in zebra finches (Taeniopyqia guttata) (Collins, ), stickleback (G.
aculeatus) (Bakker & Milinski, ) and mottled sculpin (Cottus bairdi) (Brown, ;
Downhower & Lank, ). M. L. Reid & B. D. Roitberg (cited in Roitberg et al.,
Variation in mate choice: a review 
Table .Presumed mate-choice tactics based on field studies in which female sampling behaviour was monitored
(The following are lekking species : Gallinago media,Pavo cristatus,Rupicola rupicola,Tetrao tetrix and Parotia lawesii.)
Species
N
(females) Sampling tactics
No. of males
sampled
mean (range)
Percentage
copulating with
first male Source
Fiddler crab (Uca
annulipes)
 Fixed threshold for burrow features, but
preferentially visited the burrows of
larger males
Up to  Backwell & Passmore (in press)
Great snipe (Gallinago
media)
 Pooled comparison or possibly sequential
search rule
±() Fiske & Ka
/la
/s()
Great reed warbler
(Acrocephalus
arundinaceus)
 Pooled comparison or sequential search
rule
±()Bensch & Hasslequist ()
Barnacle geese (Branta
leucopsis)
 % one-step decision rule %
sequential comparison
±() Choudhurry & Black ()
Lawes’ parotia (Parotia
lawesii)
Pooled comparison Up to  ? Pruett-Jones & Pruett-Jones
()
Pied flycatcher (Ficedula
hypoleuca)
 Pooled comparison but with repeated
visits to males}sequential comparison
±() Dale et al.()
 Pooled comparison}threshold}sequential
comparison
±()Hovi & Ra
$tti ()
Peacock (Pavo cristatus) Threshold or sequential comparison or
pooled comparison
()Petrie, Halliday & Sanders
()
Cock-of-the-rock
(Rupicola rupicola)
 Pooled comparison ±±"? Trail & Adams ()
Black grouse (Tetrao
tetrix)
 Pooled comparison ±()Rintama
$ki et al.()
Goby (Pomatoschistus
minutus)
 Threshold criteria or one-step decision
rule
±() E. Forsgren (in preparation)
"Range of means between years (N¯years).
 M. D. J  M. P
) have shown that female bark beetles Ips pini show a greater willingness to mate
with intermediate-sized males when they have first been exposed to small males than
when first exposed to large males. In the katydid Scudderia curvicauda, females showed
increased responsiveness over several playback trials, suggestive of a reduction in their
mating threshold (Tuckerman, Gwynne & Morris, ). In balloon flies (Empis
borealis), males visit swarming females. They do not appear to be able to assess absolute
female size, but can detect relative size within a swarm, preferring larger females
(Svensson & Peterson, ). Relative choice also occurs in Drosophila littoralis and D.
montana (Hoikkala & Aspi, ). Work on sex-role reversal in katydids and crickets
suggests that there is considerable plasticity in mating-decision rules in insects
(Gwynne, ). (For excellent reviews of learning in insects and potential constraints
on optimal mate choice see Papaj & Lewis, .) Other studies showing that
choosiness changes in relation to OSR, encounter rate or other factors that increase the
costs of sampling also support the claim that mating preferences are phenotypically
plastic (see Sections VI. , Section VII. ).
In some species, there is evidence for fixed mating preferences. Females may have
innate preferences for certain male phenotypes. For example, in two-choice tests, naive
virgin female guppies showed a stronger response to a male with a greater amount of
orange colouration (Brookes & Caithness, a). In cockroaches (N.cinerea) there
appear to be fixed mating thresholds for male pheromones (Moore & Moore, ). A
fixed-threshold preference for male comb size has also been reported for red junglefowl
(G.gallus) (Zuk et al., ). In the fiddler crab Uca annulipes, there also appears to be
a fixed threshold for burrow features (Backwell & Passmore, in press). In general,
however, it is hard to see how fixed preferences can persist. In poor years, many males
probably fall below a fixed threshold and if females really are inflexible they should
refuse to mate. In almost all species there is probably a time-dependent reduction in
mating threshold.
() There is considerable variability among females in the number of males sampled.
In many field studies, females often mate with the first male encountered. For example,
E. Forsgren (in preparation) found that % of female sand blennies (P.minutus)
mated with the first male encountered, while others sampled up to  males. In barnacle
geese (Branta leucopsis), % of females settled with the first mate encountered, while
others sampled up to six males (Choudhurry & Black, ). In great snipe (G.media),
% of females mated on the first observed visit to a male’s territory, others visited up
to  males (Fiske & Ka
/la
/s, ) (see Table ). It is not yet clear whether this variation
reflects differences in female receptivity, choosiness, mating-preference thresholds,
sampling tactics or is simply the result of stochastic variation (i.e. some females
encountered preferred males early in a mating sequence.
() Real () suggested that the sequential search rule is more likely to be used than
the pooled comparison when sampling costs exist (contra Janetos, ). However, in
several species where females use the pooled comparison tactic, there appear to be
substantial search costs. For example, the pooled comparison tactic has been reported
for some females in pied flycatchers (F.hypoleuca) (Dale et al., , ; Hovi
&Ra
$
tti, ), great reed warblers (Acrocephalus arundinaceus) (Bensch & Hasslequist,
) and a damselfish (Chrysiptera cyanea) (Gronell, ). The pooled comparison
tactic does seem more common in species that lek, however, and has been reported in
Variation in mate choice: a review 
most well-studied species (see Table ). This is consistent with the claim that the costs
of sampling are lower on leks, allowing a more costly sampling tactic.
()How many cues do females use ?
Many early studies analysed female choice based on the implicit assumption that
females only assessed males using a single cue. In fact, the use of multiple cues is
probably universal (e.g. Burley, ; Zuk, Ligon & Thornhill, ; Kodric-Brown,
; Choudhurry & Black, ; Borgia, ). Most empirical work shows that
behavioural traits, such as rate or intensity of display, influence female choice (reviewed
by Ryan & Keddy-Hector, ). So even when there is only one obvious morphological
ornament, choice is probably based both on display features and ornamentation (cf.
Møller & Pomiankowski,  b). Recently, theoreticians have begun to model the
evolution of multiple female preferences (Pomiankowski & Iwasa, ; Iwasa &
Pomiankowski, ; R. A. Johnstone, in preparation). Unfortunately, little is known of
the rules females use to weight the value of different signals. This is a frustrating gap
in our knowledge which can often lead to erroneous conclusions. For example, some
traits show no relationship with male mating success and are dismissed as irrelevant to
female choice. However, they may still be important to the extent that they must exceed
a threshold value (which all sexually active males fulfil) before females will consider
males as potential mates (analogous to the ante in betting ; Kodric-Brown & Brown,
; Ligon & Zwartjes,  b) (see Brookes, in press). Females may also discard or
add criteria depending on environmental conditions. For example, Thornhill ()
found that female scorpionflies (Panorpa latipennis) accepted males with prey items as
mates in a situation in which males could not always procure high-quality items.
There is some evidence that females only pay detailed attention to condition-
dependent cues that signal male quality, and that in each species only a single cue is
used (Møller & Pomiankowski,  b; Iwasa & Pomiankowski, ). This conclusion
is based on data relating to patterns of fluctuating asymmetry, the use of which as an
index of condition-dependence has been questioned (Balmford, Jones & Thomas,
b; Brookes & Caithness,  b; Jennions, in press). This finding also conflicts
with the observation that in well-studied species different investigations often find that
different traits are correlated with male ‘ attractiveness’ or mating success. For example,
in guppies (P.reticulata), tail length, display rate, parasite load, different patterns of
colouration and body size have all been implicated (Endler, ). Similarly, long-term
studies tend to show considerable between-year variation in the extent to which
different traits are correlated with mating success (Fiske et al., ).
The manner in which females utilize information from multiple cues is not clear.
There are several possibilities:
() They may treat them in a hierarchical fashion and only use lower-order cues to
discriminate between males when higher-order ones show low variation or are difficult
to discriminate (Zuk et al., ; Ligon & Zwartjes, b).
() Females may assess several cues simultaneously and give different weightings to
each. If an overall ‘index’ of attractiveness is constructed in this fashion, it raises the
intriguing possibility that males can be equally attractive in different ways. This may
provide a partial explanation for the presence of continued heritable variation in
sexually selected traits, especially if the weighting given to certain traits increases when
 M. D. J  M. P
they become less abundant resulting in frequency-dependent selection. If traits that are
more variable among males are given a higher weighting, stronger selection on these
traits may reduce variability and then lead to other traits gaining higher weighting.
With an overall ‘index’, identical female preferences can lead to females choosing males
with different values of a single cue.
() It is also possible that traits which researchers characterize as different are not
perceived as such by animals (R. C. Brookes, personal communication). For example,
researchers may speak of amount of black, orange and blue colouration as three separate
traits. It is possible, however, that these traits all contribute to net colouration, which
is what the female assesses. In the fiddler crab U.annulipes, burrows in which females
mated differed significantly from those that were sampled in six out of ten measured
variables. It is unlikely, however, that each variable was independently assessed, given
that females spend only a short amount of time in sampled burrows (e.g. Christy &
Schober, ). It is more probable that each variable contributed towards the general
suitability of the burrow (Backwell & Passmore, in press).
() Some sexual traits may not be direct choice cues but rather act as amplifiers of
variation in other cues (Hasson, ). However, in a series of interesting experiments
Brookes (in press) has shown that a trait may act as both an amplifier of another trait
and also itself be selected for. Brookes & Caithness (b,c) found that only orange
colouration was correlated with male mating success in a population of feral guppies (P.
reticulata). Although there was no correlation between black colouration and male
mating success, a preference for males with black spots was revealed when black spots
were removed by freeze-branding. Even more intriguing was that when black spots
were removed, the amount of orange colouration no longer correlated with
attractiveness to females (Brookes, in press). This suggests that melanin acts both as a
mate-choice cue and as an ‘amplifier’ of difference in orange colouration (Hasson,
). This work also illustrates the difficulties in identifying whether or not a trait is
used in mate choice when some traits (e.g. orange colouration) are more important than
others (e.g. black colouration).
To understand how females use multiple cues, we probably need to know what
information each cue provides. Several hypotheses based on information-content have
been proposed for the evolution of multiple male traits (reviewed by Møller &
Pomiankowski, b): () multiple messages, whereby each trait conveys a different
type of information about the male (e.g. parasite resistance, recent food intake,
carotenoid intake); () redundant signals: each signal on its own may not provide
sufficient information for accurate assessment, either because males can ‘cheat’ when
only one signal is involved, or because female discrimination improves when they can
combine estimates of quality from several different traits ; () unreliable signals: this
suggests that most traits do not provide information about male quality (i.e. are non-
condition-dependent Fisherian traits) and are maintained by weak female preferences
which are not particularly costly (Pomiankowski & Iwasa, ; Iwasa &
Pomiankowski, ). In support of this, Møller & Pomiankowski (b) note that
multiple ornamental traits are most often found in lekking species in which sampling
costs are probably low; () in addition to these information-based explanations, another
possibility is that different cues function at different distances. For example, in the
spotted bowerbird (Chlamydera maculata), bones spread around the bower may act to
Variation in mate choice: a review 
attract females (they are bleached white and visible from a distance). Glass nearer the
bower is encountered later and may act to stimulate the female to mate (Borgia, ).
Only when the full set of choice cues and the relative importance attached to each is
known, can we address the real extent of variation in female preferences. Being realistic,
this is probably an unattainable goal; however, studies in which two or more cues are
varied simultaneously should be conducted (see Zuk et al., ). Historically, the
emphasis has been on manipulation of a single cue while others are held constant
(reviewed in Gerhardt, ). This work has shown us which traits are potential cues
in mate choice. However, only by offering females the choice between different
combinations of traits can we test the relative importance of each cue. There is also a
strong likelihood that cues may interact in a synergistic manner. For example, female
black grouse (T.tetrix) prefer to mate with males without damaged tails and also
preferentially mate on the central territories on the lek. The effect of tail damage on
mating success varies in relation to the position of the male on the lek (Ho
$glund et al.,
). Similarly, Møller ( b) found that tail asymmetry is more strongly
discriminated against in short-tailed males than long-tailed males in barn swallows (H.
rustica). A final potential problem is that the number of cues females use may vary. For
example, females may use fewer cues when sampling costs are increased. This
explanation has been proposed to account for differences between species in the number
of ornaments (Møller & Pomiankowski,  b), but it may also account for within-
species differences when sampling behaviour shows phenotypic plasticity.
X. CONCLUSIONS
In general, sexual-selection studies have paid less attention to individual variation
among females than that among males (Rosenqvist & Berglund, ; Ahnesjo
$et al.,
). There are practical reasons for this. In many species, males are sexually active
and court for long periods, whereas females are only sexually receptive for a short time.
Males are often gaudy and conspicuous, while females are cryptic and harder to locate.
Males vary widely in their mating success, while it is less obvious how female fitness
varies in relation to mating decisions. In spite of these difficulties, we hope this review
will convince researchers that studies focusing on females as individuals rather than
the female response’ are worthy of greater attention. Studies of female sampling
behaviour may help us to understand: variation in the size and number of male
ornaments (see Sections II and IX. ) ; the maintenance of heritable variation in female
preferences (see Sections IV. and VIII); the size of the benefits provided by
choosiness (see Section II); the evolutionary history of preferences and preferred traits
(see Section IV. ); and the general design of male traits (see Section VI. ).
There is clearly still much to learn about how females choose males, and why
different females make different choices. We suggest that one profitable approach will
be to determine whether variation among females in the costs of choosiness really does
influence mate choice. It should be comparatively easy to manipulate costs (e.g.
handicapping females, manipulating parasite loads) and determine the subsequent
effects on mate choice. If true, then studies into the heritability of traits associated with
the ability to withstand costs may provide the quickest route to increased understanding
of the maintenance of heritable variation in female mating preferences. This work may
also provide information on the magnitude of the benefits associated with mate choice.
 M. D. J  M. P
Phenotypic plasticity in choosiness in relation to small changes in the costs of choice
suggest that the benefits of mate choice may sometimes be very small. Future laboratory
studies should provide more realistic choice scenarios. Many laboratory studies allow
females simultaneously to assess two or more males. In the field, sampling is usually
sequential and this may yield very different results because of memory retention (and
interference and transference, see Papaj & Lewis, ) and sampling costs.
Although we have not dealt in any detail with the neurobiology or psychology of
female choice, we suggest that this may also be a profitable area of research (reviewed
by W. Eberhard & W. Wcislo, in preparation). The general process of choice has been
extensively studied by psychologists and there is a large literature on such topics as
discrimination, learning, and memory retention, transference and interference. Most
models of mate choice have not considered these more proximate factors. This is
sometimes a problem as it may lead to unrealistic claims about what females are capable
of doing. A simple point worth keeping in mind is that most animals make many choices
about foraging every day but only rarely regarding mating. We might therefore expect
investment in the sensory apparatus and processing facilities used in feeding to be more
developed than in those used in mating. The degree of misfit between predicted and
observed choices generated by ‘ optimal foraging’ models may therefore be far smaller
than that derived from ‘optimal mate choice’ models. Of course, this argument is
affected by whether or not the rewards from the correct choice are higher for mating
than feeding decisions. Many would argue that the rewards from mate choice are high.
However, the small number of males sampled by females in many species (often the first
male encountered is accepted: see Section IX. ) and the reduction in choosiness in
response to small increases in costs (see Sections VI. &, Section VII. ) are
tantalizing pieces of evidence that this is not always the case. Mate choice for indirect
benefits may have dramatically beneficial effects in some species (e.g. Norris,  ;
Petrie ), but this is not always the case (e.g. Nicoletto, ).
Variation among females in reproductive success may be smaller than that among
males (Clutton-Brock, ); however, from a functional perspective this does not
make it any less important. For those interested in the identification of adaptations and
constraints on optimality, female mate choice and female manipulation of males is a
relatively unexplored area. The wealth of opportunities is illustrated by the fact that we
deliberately avoided reviewing variation among females in relation to multiple mating
and sperm competition. Very little is known about the extent to which females control
sperm competition (Barnett, Telford & Tibbles, ), let alone the effect of variation
in morphological, physiological and behavioural traits that influence sperm competition
or the opportunity for multiple matings. The recent publication dates for much of the
research we have cited indicates an increasing interest in the extent to which females are
active participants in sexual selection. There are exciting research opportunities ahead
for those so inclined.
XI. ACKNOWLEDGEMENTS
We thank those authors who made available their unpublished work. For useful discussions we thank Patricia
Backwell, Rob Brookes, Bill Wcislo and Bill Eberhard. M.D.J. was funded by the Rhodes Trust (Oxford).
XII. REFERENCES
A
$,I.(). Temperature affects male and female potential reproductive rates differently in the sex-role
reversed pipefish, Syngnathus typhle.Behavioural Ecology ,.
Variation in mate choice: a review 
A
$, I., V, A., A, R. V., H,T.&S,W.J.(). The role of females in
influencing mating patterns. Behavioural Ecology ,.
A, R. V., C,A.&L,A.( a). The search cost in mate choice of the pied flycatcher.
Animal Behaviour ,.
A, R. V., H
$,J.&L,A.(b). Patterns of variation in tail ornament size in birds.
Biological Journal of the Linnean Society ,.
A, R. V., H
$,J.&L,A.(). Lekking in black grouse: a test of male viability. Nature,
London ,.
A, R. V., L,A.&G,C.(). Female pied flycatchers choose territory quality and not male
characteristics. Nature,London ,.
A,G.A.&B,W.J.(). Female encounter rate and the calling behaviour and mating propensities
of male Requiena verticalis (Orthoptera : Tettigonidae). Behavioural Ecology and Sociobiology ,.
A,M.B.(). Female choice selects for extreme tail length in a widowbird. Nature,London ,
.
A,M.B.(). Sexual Selection. Princeton University Press, Princeton.
A,A.(). Female mate selection in the natterjack toad : active choice or passive attraction? Behavioural
Ecology and Sociobiology ,.
A,A.&E,M.(). Conflict, receiver bias and the evolution of signal form. Philosophical Transaction
of the Royal Society of London,Series B ,.
A, A., E,T.&R,T.(). The adaptive significance of acoustic spacing in male bushcrickets
Tettigonia viridissima: a perturbation experiment. Behavioural Ecology and Sociobiology ,.
A,S.J.&D,D.(). Animal mating systems : a synthesis based on selection theory. American
Naturalist ,.
B,P.R.Y.&P,N.I.(). Suitable approach perches affect female phonotaxis in an arboreal
frog. Herpetologica ,.
B,P.R.Y.&P, N. I. (in press). Time constraints and multiple choice criteria in the sampling
behaviour and mate choice of the fiddler crab, Uca annulipes.Behavioural Ecology and Sociobiology.
B,T.C.M.(). Genetic variation in female mating preference. Netherlands Journal of Zoology ,
.
B,T.C.M.(). Positive correlation between female preference and preferred male ornament in
sticklebacks. Nature,London ,.
B,T.C.M.&M,M.(). Sequential female choice and the previous male effect in sticklebacks.
Behavioural Ecology and Sociobiology ,.
B,T.C.M.&P,A.(). The genetic basis of female mate preferences. Journal of
Evolutionary Biology ,.
B,A.(). Mate choice on leks. Trends in Ecology and Evolution ,.
B, A., R,A.M.&A,S.D.(). Correlates of female choice in resource-defending antelope.
Behavioural Ecology and Sociobiology ,.
B, A., T,A.L.R.& J,I.L.( a). Aerodynamics and the evolution of long tails in birds.
Nature,London ,.
B, A., J,I.L.&T,A.L.R.( b). On avian asymmetry: evidence of natural selection for
symmetrical tails and wings in birds. Proceedings of the Royal Society of London,Series B ,.
B, G. W., F,R.C.&B,J.V.(). Do the colour of parents, companions and self
influence assortative mating in the polychromatic Midas cichlid? Animal Behaviour ,.
B,P.(). Ornaments and body size variation in some African passerine birds. Doctoral thesis, Uppsala
University, Sweden.
B, M., T,S.R.&T,B.J.(). Female mediation of sperm competition in the millipede
Alloporus uncinatus (Diplopoda: Spirostreptidae). Behavioural Ecology and Sociobiology ,.
B,P.(). Optimal outbreeding. In Mate Choice (ed. P. Bateson), pp. . Cambridge University
Press, Cambridge.
B, A. T. D., C,I.C.&N,K.J.(). Sexual selection and the mismeasure of colour.
American Naturalist ,.
B,S.&H,D.(). Evidence for active female choice in a polygynous warbler. Animal
Behaviour ,.
B,A.(). Risky sex : male pipefishes mate at random in the presence of a predator. Animal Behaviour
,.
B,A.(). The operational sex ratio influences choosiness in a pipefish. Behavioural Ecology ,.
 M. D. J  M. P
B, S., H,D.&W,I.(). A theory of fads, fashion, custom and cultural changes
as information cascades. Journal of Political Economics ,.
B,T.R.&M,A.P.(). Sperm Competition in Birds. Academic Press, New York.
B, P. J., J,M.D.&P,N.I.(). Chorus size and call intensity : female choice in the
painted reed frog. Hyperolius marmoratus.Behaviour ,.
B,C.R.B.(). Repeatability : its role in evolutionary studies of mating behaviour. Evolutionary Ecology
,.
B,G.(). The cost of display in the non-resource based mating system of the satin bowerbird. American
Naturalist ,.
B,G.(). Complex male display and female choice in the spotted bowerbird: specialised functions for
different bower decorations. Animal Behaviour ,.
B,G.R.(). Proximate costs and benefits of mate acquisition at leks of the frog Ololygon rubra.Animal
Behaviour ,.
B,J.W.(). The evolution of leks. In Natural Selection and Social Behaviour (ed. R. D. Alexander and
D. W. Tinkle), pp. . Chiron Press, New York.
B,J.W.&A, M. B. (eds) (). Sexual Selection: Testing the Alternatives. John Wiley and
Sons, New York.
B, J. W., V,S.L.&G,R.(). Leks and the unanimity of female choice. In
Evolution: Essays in Honour of John Maynard Smith (ed. P. J. Greenwood, P. H. Harvey and M. Slatkin), pp.
. Cambridge University Press, Cambridge.
B, F., G,H.C.&B,R.K.(). Female choice and genetic correlations. Trends in Ecology
and Evolution ,.
B,F.&H,K.(). A test of indirect models of selection in the Trinidad guppy. Heredity ,
.
B, R. C. (in press). Naturwissenschaften.
B,R.C.&C,N.( a). Does a male’s attractiveness to a female depend on her previous
experience? South African Journal of Science ,.
B,R.C.&C,N.( b). Female choice in a feral guppy population : are there multiple cues?
Animal Behaviour ,.
B,R.&C,N.( c). Manipulating a seemingly non-preferred male ornament reveals a role in
female choice. Proceedings of the Royal Society of London,Series B ,.
B,L.(). Patterns of female choice in mottled sculpins (Cottidae : Teleostei). Animal Behaviour ,
.
B,W.D.(). Size-assortative mating in the blister beetle Lytta magister (Coleoptera : Meloidae) is due
to male and female preference for larger mates. Animal Behaviour ,.
B,W.D.(). The cause of size-assortative mating in leaf beetle Trirhabda canadensis (Coleoptera:
Chrysomelidae). Behavioural Ecology and Sociobiology ,.
B,N.(). Mate choice by multiple criteria in a monogamous species. American Naturalist ,.
B,N.(). The meaning of assortative mating. Ethology and Sociobiology ,.
B,N.(). The differential allocation hypothesis: an experimental test. American Naturalist ,.
B,R.K.(). The variability of mating signals and preferences in the brown planthopper, Nilaparvata
lugens (Homoptera: Delphacidae). Journal of Insect Behaviour ,.
B,R.K.(). Genetic variation in mating signals and responses. In Speciation and the Recognition
Concept: Theory and Application (ed. D. M. Lambert and H. G. Spencer). Johns Hopkins University,
Baltimore.
C,C. &B,P.(). Sexual selection: the evolution of conspicuous characteristics in birds by
means of imprinting. Evolution ,.
C,C. &B,P.(). Sexual imprinting and a preference for ‘ supernormal’ partners in Japanese
quail. Animal Behaviour ,.
C, M., B,R.K.&H,G. M.(). Genetic variation in song and female song preference
in the grasshopper Chorthippus brunneus (Orthoptera : Acrididae). Animal Behaviour ,.
C,M.I.(). Tail length and female choice. Trends in Ecology and Evolution ,.
C,S.&B,J.M.(). Mate-selection behaviour and sampling strategies in geese. Animal
Behaviour ,.
C,S.&B,J.M.(). Barnacle geese preferentially pair with familiar associates from early life.
Animal Behaviour ,.
C,J.H.(). Mimicry, mate choice and the sensory trap hypothesis. American Naturalist ,.
C,J.H.&B,P.R.Y.(). Sensory exploitation. Trends in Ecology and Evolution ,.
Variation in mate choice: a review 
C,J.H.&S,U.M.(). A test for resource-defence mating in the fiddler crab Uca beebei.Animal
Behaviour ,.
C-B, T. H. (ed.) (). Reproductive Success. University of Chicago Press, Chicago.
C-B, T. H., D,J.C.&N,R.J.C.(). The evolution of ungulate leks. Animal
Behaviour ,.
C-B, T. H., G, D., H-H,M.&A,S.D.(). Passing the buck: resource
defence, lek breeding and mate choice in fallow deer. Behavioural Ecology and Sociobiology ,.
C-B, T. H., H-H,M.&R,A.(). Mate choice on fallow deer leks.
Nature,London ,.
C-B,T.H.&P,G.A.(). Potential reproductive rates and the operation of sexual selection.
Quarterly Review of Biology ,.
C-B, T. H., P,O.&MC,A.(). Mate retention, harassment and the evolution of
ungulate leks. Behavioural Ecology ,.
C,J.A.(). Sexual selection and the psychophysics of female choice. Zeitschrift fuXr Tierpsychologie ,
.
C,R.D.&C
!,R.T.(). Selection for increased pheromonal response in the male pink bollworm,
Pectinophora gossypiella (Lepidoptera: Gelechiidae). Behavioural Genetics ,.
C,S.A.(). The effect of recent experience on female choice in zebra finches. Animal Behaviour ,
.
C
#,I.M.&H,W.(). Male and female mate choice in the redlipped blenny. Animal Behaviour ,
.
C,H.(). The Ant and the Peacock. Cambridge University Press, Cambridge.
C,S.A.(). Changes in mating behaviour produced by selection for ethological isolation between
ebony and vestigial mutants of Drosophila melanogaster.Evolution ,.
C, P. H., T, S. E., L, M. C., C, S. L., S,A.S.&S,C. R.(). Mate density,
predation risk and the seasonal sequence of mate choices: a dynamic game. American Naturalist ,.
D, S., A, T., L,J.T.&S,T.(). Mate sampling behaviour of female pied
flycatchers: evidence for active mate choice. Behavioural Ecology and Sociobiology ,.
D, S., R,H.&S,T.(). Competition for a mate restricts mate search of female pied
flycatchers. Behavioural Ecology and Sociobiology ,.
D,S.&S,T.(). Female contests for nest sites and mates in the pied flycatcher Ficedula
hypoleuca.Ethology ,.
D,N.B.(). Mating systems. In Behavioural Ecology rd ed. (ed. J. R. Krebs and N. B. Davies), pp.
. Blackwell Scientific Publications, Oxford.
D,M.S.&G,T.(). An exaggerated preference for simple neural models of signal evolution ?
Proceedings of the Royal Society of London,Series B ,.
 L,F.&M,A.P.(). Female reproductive effort depends on the degree of ornamentation of their
mate. Evolution ,.
D,J.C.&N,R.J.C.(). Olfactory cues influence female choice in two lek-breeding antelopes.
Nature,London ,.
D,J.C.&W,P.(). Uganda kob prefer high-visibility leks and territories. Behavioural Ecology ,
.
 W,A.J.(). The genetic basis and evolution of acoustic mate recognition signals in a Ribautodelphax
planthopper (Homoptera, Delphacidae). . The female call. Journal of Evolutionary Biology ,.
D,Y.&P,N.(). On adaptive search and optimal stopping in sequential mate choice.
American Naturalist ,.
D,J.F.&L,D. B.(). Effect of previous experience on mate choice by female mottled sculpins.
Animal Behaviour ,.
D,L.A.(). Sexual selection and imitation: females copy the mate choice of others. American
Naturalist ,.
D,L.A.&G, J.-G. J. (). Female mate copying in the guppy (Poecilia reticulata) : age-dependent
effects. Behavioural Ecology ,.
D, M. L., H,S.P.&P,N.I.(). The effect of changes in the relative timing of signals during
female phonotaxis in the painted reed frog. Hyperolius marmoratus.Animal Behaviour ,.
D,M.L.&P,N.I.(). The combined effect of intensity and the temporal relationship of
stimuli on phonotaxis in female painted reed frogs. Animal Behaviour ,.
D, M. L., P, N. I., B,P.J.&H,S.P.(). Male behaviour and correlates of mating
success in a natural population of African painted reed frog (Hyperolius marmoratus). Herpetologica ,.
 M. D. J  M. P
E,W.&W, W. (in preparation). Typology, sexual selection by female choice and neural systems.
E,M.&P,R.(). Inter-sexual conflict over copulations in the European starling : evidence for the
female mate-guarding hypothesis. Behavioural Ecology and Sociobiology ,.
E,M.&P, R. (In press). Female European starlings increase their copulation solicitation rate when
faced with the risk of polygyny. Animal Behaviour .
E, M., P,R.&K,B.(). Do female tree swallows guard their mates by copulating
frequently? A comment on Whittingham et al. Animal Behaviour ,.
E,D.J.(). Environmental control of horn length dimorphism in the beetle Onthophagus acuminatus
(Coleoptera). Proceedings of the Royal Society of London,Series B ,.
E,J.(). Multiple trait coevolution and environmental gradients in guppies. Trends in Ecology and
Evolution ,.
E,J.A.&H,A.E.(). Geographic variation in female preferences for male traits in Poecillia
reticulata.Evolution ,.
E, G., F,S.P.&D,T.H.(). Genetic differences in mating success and female choice in
seaweed flies (Coelopa frigida). Heredity ,.
E,M.&A,A.(). Selection for exaggerated male traits by female aesthetic preferences. Nature,
London ,.
E,M.R.&T, A.L.R. (). The aerodynamics and mechanical effects of elongated tails in the
scarlet-tufted malachite sunbird : measuring the costs of a handicap. Animal Behavioural ,.
F,R.A.(). The Genetical Theory of Natural Selection. Clarendon Press, Oxford.
F,P.&K
/
/,J.A.(). Mate sampling and copulation behaviour of great snipe females. Animal Behaviour
,.
F, P., K
/
/,J.A.&S,S.A.(). Correlates of male mating success in the lekking great snipe
(Gallinago media). Behavioural Ecology ,.
F,D.J.C. &M, C. D. (eds) (). Kin Recognition in Animals. Wiley, Chichester.
F,T.G.&R,R.(). Models of female choice in acoustic communication. Behavioural Ecology ,
.
F,E.(). Predation risk affects mate choice in a gobiid fish. American Naturalist ,.
F, E. (in preparation). Mate sampling in a population of sand gobies. Pomatoschistus minutus.
F, C. W., ML,L.A.&M,T.A.(). Male body size affects female lifetime reproductive
success in a seed beetle. Animal Behaviour ,.
G,H. C.(). Vocalisations of some hybrid treefrogs : acoustic and behavioural analyses. Behaviour ,
.
G,H.C.(). Sound pattern recognition in some North American treefrogs (Anura: Hylidae):
implications for mate choice. American Zoologist ,.
G,H.C.(). Evolutionary and neurobiological implications of selective phonotaxis in the green
treefrog. Hyla cinerea.Animal Behaviour ,.
G,H.C.(). Conducting playback experiments and interpreting their results. In Playbacks and
Studies of Animal Communication (ed. P. K. McGregor). Plenum Press, New York.
G,H.C.&K,G.M.(). Masking of acoustic signals by the chorus background noise in the
green tree frog : a limitation on mate choice. Animal Behaviour ,.
G,T.(). Search, discrimination, and selection : mate choice by pied flycatchers. American Naturalist ,
.
G,R.M.(). Field playback of male display attracts females in lek-breeding sage grouse. Behavioural
Ecology and Sociobiology ,.
G,R.M.&B,G.C.(). The costs of female choice in a lekking bird. Behavioural Ecology ,
.
G,R.M.&B,J.W.(). Lek organization in sage grouse : variation on a territorial theme. Auk
,.
G, R. M., B,J.W.&V,S.L.(). Mate choice in lekking sage grouse revisited : the
role of vocal displays, female site fidelity and copying. Behavioural Ecology ,.
G,R.M.&H
$,J.(). Copying and sexual selection. Trends in Ecology and Evolution ,.
G,A.S.&D,T.H.(). Evolution of female choice in seaweed flies : fisherian and good gene
mechanisms operate in different populations. Proceedings of the Royal Society of London,Series B ,.
G, A. S., F,S.P.&D,T.H.(). Genetic correlation between a female mating preference and
the preferred male character in seaweed flies (Coelopa frigida). Evolution ,.
G, J.-G. J. & D,L.A.(). Variability and repeatability of female mating preferences in the
guppy. Animal Behaviour ,.
Variation in mate choice: a review 
G,R.(). Limited female choice and the evolution of secondary sexual characters. American
Naturalist ,.
G,A.(). Biological signals as handicaps. Journal of Theoretical Biology ,.
G,M.D.(). Cooperation and conflict in the evolution of signal interactions. Annual Reviews of
Ecology and Systematics ,.
G,A.M.(). Visiting behaviour by females of the sexually dichromatic damselfish, Chrysiptera cyanea
(Teleostei: Pomacentridae): a probable method of assessment. Ethology ,.
G,D.T.(). Sex-biased predation and the risky mate-locating behaviour of male tick-tock cicadas
(Homoptera: Cicadidae). Animal Behaviour ,.
G,D.T.(). Sexual competition among females : what causes sex-role reversal? Trends in Ecology and
Evolution ,.
G,D.T.&B,W.D.(). Mate feeding, offspring investment, and sexual differences in katydids
(Orthoptera: Tettigonidae). Behavioural Ecology ,.
G,D.T.&S,L.W.(). Experimental reversal of courtship roles in an insect. Nature,London
,.
H,S.E.&G,J.L.(). Female platys prefer long tails. Nature,London ,.
H,O.(). Sexual displays as amplifiers : practical examples with an emphasis on feather decorations.
Behavioural Ecology ,.
H,A.V.(). Female choice and the heritability of attractive male traits : an empirical study. American
Naturalist ,.
H,A.V.&D,L.M.(). Mate choice by female crickets is influenced by predation risk. Animal
Behaviour ,.
H,I.L.(). Inheritance of female mating propensities for yellow locus genotypes in Drosophila
melanogaster.Genetical Research,Cambridge ,.
H, I. L., A, M. B., A, S. T., B, C. R., B, G., H, G., K, M.,
L, R., M S, J., O’D, P., T,R.&W,F.J.(). The evolution of
mating preferences and sexually selected traits. In Sexual Selection : Testing the Alternatives (ed. J. W. Bradbury
and M. B. Andersson), pp. . John Wiley & Sons, New York.
H,G.E.(). Plumage colouration is a sexually selected indicator of male quality. Nature,London ,
.
H
$,J.&A,R.V.(). Leks. Princeton University Press, Princeton.
H
$, J., A, R. V., G,R.M.&L,A.(). Mate-choice copying in black grouse.
Animal Behaviour ,.
H
$, J., A, R. V., L,A.&R
$
,O.(). Context-dependent effects of tail-ornament
damage in black grouse. Behavioural Ecology ,.
H,A.A.&A,J.(). Criteria of female mate choice in Drosophila littoralis.D.montana and D.
ezoana.Evolution ,.
H,A.E.(). Genetic differentiation in female choice between two guppy populations. Animal Behaviour
,.
H,A.E.(). Evolution by sexual selection: what can population comparisons tell us? American Naturalist
,.
H,A.E.(). Effect of artificial selection on male colour patterns on mating preference of female guppies.
Proceedings of the Royal Society of London,Series B ,.
H,A.E.&E,J.A.(). Correlated evolution of female mating preferences and male color patterns
in the guppy Poecilia reticulata.Science,Washington ,.
H,A.M.(). Female zebra finches choose extra-pair copulations with genetically attractive males.
Proceedings of the Royal Society of London,Series B ,.
H,M.&R
$
,O.(). Mate sampling and assessment procedures in female flycatchers (Ficedula
hypoleuca). Ethology ,.
H,S.P.&J,L.K.(). Environmental variance in lifetime mating success, mate choice and
sexual selection. American Naturalist ,.
H, F. M., P, M., O, M., B,T.R.&M,A.P.(). Why do females copulate
repeatedly with one male ? Trends in Ecology and Evolution ,.
H, J. M. C., MN,J.M.&C,I.C.(). Song, sexual selection, starvation and strategic
handicaps. Animal Behaviour ,.
I,H.&M,O.(). Directional selection for pulse repetition rate of the courtship sound and correlated
responses occurring in several characters in Drosophila mercatorum.Japanese Journal of Genetics ,.
 M. D. J  M. P
I,Y.&P,A.(). The evolution of mate preferences for multiple sexual ornaments.
Evolution ,.
I,Y.&P,A.(). Continual changes in mate preferences. Nature,London ,.
I, Y., P,A.&N,S.(). The evolution of costly mate preferences. II. The handicap
principle. Evolution ,.
J,A.C.(). Strategies of female mate choice : a theoretical analysis. Behavioural Ecology and
Sociobiology ,.
J,M.D.(). Female choice in birds and the cost of long tails. Trends in Ecology and Evolution ,
.
J,M.D.(). Causes of cricket synchrony. Current Biology ,.
J, M. D. (in press). The allometry of fluctuating asymmetry in Southern African plants: flowers and
leaves. Biological Journal of the Linnean Society.
J, M. D., B,P.R.Y.&P,N.I.(). Repeatability of mate choice : the effect of size
in the African frog, Hyperolius marmoratus.Animal Behaviour ,.
J,L.K.(). Sexual selection in a brentid weevil. Evolution ,.
J, K., T, R., L,J.D.&Z,M.(). The direction of mothers’ and daughters’
preferences and the heritability of male ornaments in red jungle fowl (Gallus gallus). Behavioural Ecology ,
.
J,R.A.(). Honest signalling, perceptual error and the evolution of all-or-nothing displays.
Proceedings of the Royal Society of London,Series B ,.
J,R.A.(). Sexual selection, honest advertising and the handicap principle: reviewing the evidence.
Biological Reviews of the Cambridge Philosophical Society ,.
J,I.L.&H,F.M.(). Mutual sexual selection in a monogamous seabird. Nature,London ,
.
K,K.Y.(). The dynamics of sexual selection and founder effects in species formation. In Genetics,
Speciation and the Founder Principle (ed. L. V. Giddings, K. Y. Kaneshiro and W. W. Anderson), pp. .
Oxford University Press, New York.
K, P. W. E., T, I. P. M., V, C. J. & O’D,P.(). Non-random mating in Adalia
bipunctata (the two spot ladybird). II. Further tests for female mating preferences. Heredity ,.
K,B.(). Polygyny in the blue tit : unbalanced sex ratio and female aggression restrict mate choice.
Animal Behaviour ,.
K,B.(). Polygyny in the blue tit: intra- and inter-sexual conflict. Animal Behaviour ,.
K, B., V, G. R.,   B, M., B, T.,  B,C.&D,A.A.
(). Extra-pair paternity results from female preference for high quality males in the blue tit. Nature,London
,.
K,M.&B,N.(). De
!ja
'vu all over again. Nature,London ,.
K,M.&D,L.A.(). Sexual selection and the evolutionary effect of copying mate choice.
Behavioural Ecology and Sociobiology ,.
K,M.&R,M.J.(). The paradox of the lek and the evolution of mating preferences. Nature,
London ,.
K-B,A.(). Female choice of multiple male criteria in guppies : interacting effects of dominance,
colouration and courtship. Behavioural Ecology and Sociobiology ,.
K-B,A.&B,J.H.(). Truth in advertising: the kinds of traits favoured by sexual selection.
American Naturalist ,.
K,H.R.(). Selection for sexual isolation between geographic forms of Drosophila mojavensis. II.
Effects of selection on mating preferences and propensity. Evolution ,.
L,R.(). Models of speciation by sexual selection of polygenic traits. Proceedings of the National Academy
of Sciences,U.S.A.,.
L,W.S.(). Mate choice and competition in Tetraopes tetraophthalamus : effects of local sex ratio
variation. Behavioural Ecology and Sociobiology ,.
L,J.T.&S,T.(). Female pied flycatchers Ficedula hypoleuca choose male characteristics in
a homogenous environment. Behavioural Ecology and Sociobiology ,.
L,J.D.&Z,P.W.( a). Female red junglefowl choose to mate with multiple males. Animal
Behaviour ,.
L,J.D.&Z,P.W.( b). Ornate plumage of male red junglefowl does not influence mate choice
by females. Animal Behaviour ,.
L,A.(). Sexual behaviour of the lek-forming white-bearded manakin (Manacus manacus trinitatis).
Zeitschrift fuXr Tierpsychologie ,.
Variation in mate choice: a review 
L,A.(). Lek behaviour in the golden-headed manakin (Pipra erythrocephalus) in Trinidad (West Indies).
Fortschr. Verhallensforsch. Heft . Paul Parey, Berlin, Germany.
L,G.S.J,S, F. G., T, T. M., T, W. A. III. & T Z  G S
C (). Copying others, an evolutionary stable strategy for mate choice: a model. American Naturalist ,
.
L,J.R.&H,R.D.(). Alternative reproductive tactics in anurans, dynamic games with density
and frequency dependence. American Naturalist ,.
M, T., S, R., L,J.&H
/
,T.(). Why do female adders copulate so frequently ?
Nature,London ,.
M,C.(). Predation as a cost of reproduction. Trends in Ecology and Evolution ,.
M, M. E. N., O’D,P.&W,J.(). Female mating preference is genetic. Nature,London ,
.
M, M. E. N., O’D, P., K,P.W.E.&I,H.(). Genetics and the evolution of female
choice. Nature,London ,.
M,J.H.(). Male aggression: a cost of female mate choice in Cayo Santiago rhesus monkeys. Animal
Behaviour ,.
M,K.(). Dark habitats and bright birds illustrate the role of the environment in species divergence.
Nature,London ,.
MC,K.&C-B,T.(). Is mate choice copying or aggregation responsible for skewed
distributions of females on leks. Proceedings of the Royal Society of London,Series B ,.
M,M.&B,T.C.M.(). Female sticklebacks use male colouration in mate choice and hence
avoid parasitized males. Nature,London ,.
M,M.&B,T.C.M.(). Costs influence sequential mate choice in stickleback, Gasterosteus
aculeatus.Proceedings of the Royal Society of London,Series B ,.
M,R.L.&G,M.D.(). Psychoacoustics of female phonotaxis and the evolution of signal
interactions in orthoptera. Ethology,Ecology and Evolution ,.
M,A.P.(). Female choice selects for male sexual tail ornaments in the monogamous swallow. Nature,
London ,.
M,A.P.( a). Frequency of female copulations with multiple males and sexual selection. American
Naturalist ,.
M,A.P.(b). Female swallow preference for symmetrical male sexual ornaments. Nature,London ,
.
M,A.P.( a). Sexual Selection and the Barn Swallow. Oxford University Press, Oxford.
M,A.P.( b). Repeatability of female choice in a monogamous swallow. Animal Behaviour ,.
M,A.P.( c). Symmetrical male sexual ornaments, paternal care, and offspring quality. Behavioural
Ecology ,.
M,A. P.& L,F.(). Differential costs of a secondary sexual character: an experimental test of the
handicap principle. Evolution ,.
M,A.P.&P,A.( a). Fluctuating asymmetry and sexual selection. Genetica ,.
M,A.P.&P,A.( b). Why have birds got multiple sexual ornaments? Behavioural Ecology
and Sociobiology ,.
M,A.J.(). S   Nauphoeta cinerea: inherited mating preferences? Behavioural Genetics
,.
M,A.J.(). The inheritance of social dominance, mating behaviour and attractiveness to mates in male
Nauphoeta cinerea.Animal Behaviour ,.
M,A.J.(). Genetic evidence for the ‘good genes’ process in sexual selection. Behavioural Ecology and
Sociobiology ,.
M,A.J.&M,P.J.(). Female strategy during mate choice : threshold assessment. Evolution ,
.
M,M.R.(). Female choice of large males in the treefrog. Hyla chrysoscelis : the importance of identifying
the scale of choice. Behavioural Ecology and Sociobiology ,.
N,R.A.&B,R.K.(). Does runaway sexual selection occur in finite populations ? Journal of
Evolutionary Biology ,.
N,P.F.(). Offspring quality and female choice in the guppy. Poecillia reticulata.Animal Behaviour
,.
N
$,R.&H,P.(). Biological markets : supply and demand determine the effect of partner choice
in cooperation, mutualism and mating. Behavioural Ecology and Sociobiology ,.
 M. D. J  M. P
N,K.(). Heritable variation in a plumage indicator of viability in male great tits Parus major.Nature,
London ,.
O’D,P.&M,M.(). Non-random mating in Adalia bipunctata (the two-spot ladybird). III. New
evidence of a genetic preference. Heredity ,.
O,M.&M,T.(). Female choice on male quantitative traits in lizards : why is it so rare?
Behavioural Ecology and Sociobiology ,.
O, I. F. P., B,T.&T,D.B. A.(). Extraordinary sex roles in the Eurasian dotterel : female
mating arenas, female–female competition, and female mate choice. American Naturalist ,.
P, P., A,R.V.&K,E.(). Female choice in the kestrel under different availability
of mating options. Animal Behaviour ,.
P,D.R.&L, A. C. (eds) (). Insect Learning : Ecological and Evolutionary Perspectives. Chapman &
Hall, New York.
P,G.A.(). Evolution of competitive mate searching. Annual Reviews of Entomology ,.
P,G.A.(). Sexual selection and sexual conflict. In Sexual Selection and Reproductive Competition in
Insects (ed. M. S. Blum and N. A. Blum), pp. . Academic Press, New York.
P,G.A.(). Mate quality and mating decisions. In Mate Choice (ed. P. Bateson), pp. .
Cambridge University Press, Cambridge.
P,M.(). Mate choice in role-reversed species. In Mate Choice (ed. P. Bateson), pp. . Cambridge
University Press, Cambridge.
P,M.(). Mating decisions by common moorhens (Gallinula chloropus). In Acta XIX Congressus
Internationalis Ornithologici (ed. H. Ouellet). University of Ottawa Press, Ottawa.
P,M.(). Copulation frequency in birds : why do females copulate more than once with the same male ?
Animal Behaviour ,.
P,M.(). Improved growth and survival of offspring of peacocks with more elaborate trains. Nature,
London ,.
P, M., H, M., H, T. R., B,H.&P,C.(). Multiple mating in a lekking bird :
why do peahens mate with more than one male and with the same male more than once ? Behavioural Ecology
and Sociobiology ,.
P, M., H,T.&S,C.(). Peahens prefer peacocks with elaborate trains. Animal Behaviour
,.
P,M.&H,F.M.(). Intraspecific variation in courtship and copulation frequency : an effect of
mismatch in partner attractiveness. Behaviour ,.
P,M.&J, M. D. (in press). Sperm competition : a female perspective. In Sperm Competition
(ed. T. R. Birkhead and A. P. Møller). Academic Press.
P,R.&D,L.M.(). Predation on females or males: who pays for bright male traits ? Animal
Behaviour ,.
P, D. A., B, P. R. Y., C,N.&J,M.D.(). Stabilising or directional
selection in signalling systems : investigations in a population of painted reed frogs. Hyperolius marmoratus.
South African Journal of Science ,.
P,A.(). The evolution of female mating preferences for male genetic quality. Oxford Surveys
in Evolutionary Biology ,.
P,A.&I,Y.(). Evolution of multiple sexual preferences by Fisher’s runaway process of
sexual selection. Proceedings of the Royal Society of London,Series B ,.
P,A.&M,A.P.(). A resolution of the lek paradox. Proceedings of the Royal Society of
London,Series B ,.
P,A.&S,L.(). Female choice and genetic correlations. Trends in Ecology and Evolution
,.
P, W. K., M,C.J.&W,E.K.(). Mating patterns in semi-natural populations of mice
influenced by MCH genotype. Nature,London ,.
P,R.(). Mate choice decisions by parasitized upland bullies, Gobiomorphus breviceps.Proceedings of the
Royal Society of London,Series B ,.
P,H.C.(). Courtship in the water mite Neumania papillator : males capitalize on female adaptations
for predation. Animal Behaviour ,.
P-J,S.G.(). Independent versus nonindependent mate choice : do females copy each other?
American Naturalist ,.
P-J,S.G.&P-J,M.A.(). Sexual selection through female choice in Lawes’ parotia, a
lek-mating bird of paradise. Evolution ,.
Variation in mate choice: a review 
P,M.(). Mating patterns and a test for female mate choice in Etheostomata spectabile (Pisces, Percidae).
Behavioural Ecology and Sociobiology ,.
R,L.A.(). Search theory and mate choice. I. Models of single-sex discrimination. American Naturalist
,.
R, L. A. (ed.) (). Behavioural Mechanisms in Evolutionary Ecology. University of Chicago Press, Chicago.
R,M.L.&W,P. J.(). Mate-choice criteria of Ipswich sparrows : the importance of variability.
Animal Behaviour ,.
R,J. D.&C
#,I. M.(). Direct selection on mate choice : female redlip blennies pay more for better
males. Behavioural Ecology ,.
R,J.D.&G,M.R.(). Costs and benefits of female mate choice : is there a lek paradox? American
Naturalist ,.
R,J.D.&G,M.R.(). Female mate preference enhances offspring growth and reproduction in
a fish, Poecilia reticulata.Proceedings of the Royal Society of London,Series B ,.
R
$, P. T., A, R. V., H
$,J.&L,A.(). Mate sampling behaviour of black
grouse females (Tetrao tetrix). Behavioural Ecology and Sociobiology ,.
R,M.G.(). Female preferences for ‘ song races’ of Ephippiger ephippiger (Orthoptera: Tettigoniidae).
Animal Behaviour ,.
R,M.G.( a). Setbacks in the search for mate-preference genes. Trends in Ecology and Evolution ,
.
R,M.G.( b). Variation in male song and female preferences within a population of Ephippiger
ephippiger (Orthoptera: Tettigoniidae). Animal Behaviour ,.
R,J.G.M. (). Female choice, male strategies and the role of vocalizations in the Australian frog,
Uperolia rugosa.Animal Behaviour ,.
R, J.G.M. (). Female choice increases fertilization success in the Australian frog, Uperolia
laevigata.Animal Behaviour ,.
R, W. L., D, J.-W., L, C., G,T.J.&B,L.B.(). The potential for genetic
manipulation of the redbanded leafroller moth sex pheromone blend. In Evolutionary Genetics of Invertebrate
Behaviour: Progress and Prospects (ed. M. D. Huttell), pp. . Plenum Press, New York.
R, B. D., R,M. L.&L,C.(). Choosing hosts and mates : the value of learning. In Insect Learning:
Ecological and Evolutionary Perspectives (ed. D. R. Papaj and A. C. Lewis), pp. . Chapman & Hall, New
York.
R,G.&B,A.(). Is female sexual behaviour a neglected topic ? Trends in Ecology and
Evolution ,.
R, L., A, G., S,A.&K,J.(). Sexual conflict and the evolutionary ecology of mating
patterns: water striders as a model system. Trends in Ecology and Evolution ,.
R, W. J., B, K. J., J,J.J.&F,J.(). Video playback experiments on stickleback
mate choice: female motivation and attentiveness to male colour cues. Animal Behaviour ,.
R,M.J.(). The Tungara Frog. University of Chicago Press, Chicago.
R,M.J.(). Mechanisms underlying sexual selection. In Behavioural Mechanisms in Evolutionary Ecology
(ed. L. A. Real), pp. . University of Chicago Press, Chicago.
R, M. J., F, J. H., W,W.&R,A.S.(). Sexual selection for sensory exploitation in the
frog Physalaemus pustulosus.Nature,London ,.
R,M.J.&K-H,A.(). Directional patterns of female mate choice and the role of sensory biases.
American Naturalist ,S–S.
R, M. J., P,S.A.&W,W.(). Auditory tuning and call frequency predict population-
based mating preferences in the cricket frog, Acris crepitans.American Naturalist ,.
R,M.J.&W,W.E.(). Asymmetries in mating preferences between species : female swordtails
prefer heterospecific males. Science,Washington ,.
R,M.J.&W,W.(). Coevolution of sender and receiver : effect on local mate preference in
cricket frogs. Science,Washington ,.
S,T.W.&T,O.R.(). Disruptive sexual selection in Colias eurytheme butterflies.
Proceedings of the National Academy of Sciences,U.S.A.,.
S, I., M,C.A.&R,M.J.(). Benefits to male sailfin mollies of mating with heterospecific
females. Science,Washington ,.
S,J.J.(). The function of call alternation in anuran amphibians : a test of three hypotheses. Evolution
,.
S,J.J.(). Male advertisement and female choice in frogs: recent findings and new approaches to the
study of communication in a dynamic acoustic environment. American Zoologist ,.
 M. D. J  M. P
S,J.J.&G,H.C.(). Spatially mediated release from auditory masking in an anuran
amphibian. Journal of Comparative Physiology A ,.
S,K.(). Phylogenetic tests of the sensory exploitation model of sexual selection. Trends in Ecology and
Evolution ,.
S,T.E.&B,W.J.(). Experimental manipulation of mate choice by male katydids : the effect of
female encounter rate. Behavioural Ecology and Sociobiology ,.
S,S.M.(). Courtship and female mate selection in a marine isopod crustacean. Paracerceis sculpta.
Animal Behaviour ,.
S, A., K,J.&T,S.(). An experimental study on the effects of predation risk and feeding regime
on the mating behaviour of the water strider. American Naturalist ,.
S,L.W.(). Kin recognition and its influence on mating preferences of the field cricket, Gryllus
bimaculatus (de Geer). Animal Behaviour ,.
S,L.W.(). Courtship role reversal in bush crickets: another role for parasites? Behavioural Ecology
,.
S,T.&D,S.(). Why do female pied flycatchers mate with already mated males: deception or
restricted mate sampling? Behavioural Ecology and Sociobiology ,.
S,T.&L,J.T.(). Polygyny in birds: the role of competition between females for male
parental care. American Naturalist ,.
S, T., L, T. J., S,G.&B,T.(). On the cost of searching for a mate in
female pied flycatchers Ficedula hypoleuca.Animal Behaviour ,.
S,K.&M, A.-M. (). Female mating behaviour in the field cricket (Gryllus pennsylvanicus)
(Orthoptera: Gryllidae) at different operational sex ratios. Journal of Inset Behaviour ,.
S, R., C-B,T.H.&S,W.J.(). Black holes, mate retention and the evolution
of ungulate leks. Behavioural Ecology ,.
S, P., S, J. B., MD,D.W.&J,C.S.(). Female multi-mating in the common
shrew as a strategy to reduce inbreeding. Proceedings of the Royal Society of London,Series B ,.
S,B.K.(). Passive and active female choice : a comment. Animal Behaviour ,.
S,M.S.(). Assessing female choice for mates when the males’ characters vary during the sampling
period. Animal Behaviour ,.
S,M.S.(). Mate choice as an information gathering process under time constraints : implications for
behaviour and signal design. Animal Behaviour ,.
S,K.(). Sexual selection and intra-female competition in the green poison dart frog. Dendrobates
auratus.Animal Behaviour ,.
S,K.(). Parental care and the cost of polygyny in the green dart-poison frog, Dendrobates auratus.
Behavioural Ecology and Sociobiology ,.
S,K.(). Mating strategies in two species of dart-poison frogs : a comparative study. Animal Behaviour
,.
S,B.G.&P,E.(). Mate choice tactics and swarm size : a model and a test in a dance fly.
Behavioural Ecology and Sociobiology ,.
T,G.&T,J.M.(). Reversal of mating preference by crossing strains of Drosophila melanogaster.
Nature,London ,.
T,S.R.(). Mechanisms and evolution of inter-male spacing in the painted reedfrog (Hyperolius
marmoratus). Animal Behaviour ,.
T, S. R., D,M.L.&P,N.I.(). Mate choice occurs only in small choruses of painted
reed frogs Hyperolius marmoratus.Bioacoustics ,.
T,S.(). Male spawning success and female choice in the mottled triplefin. Forsterygion varium
(Pisces: Tripterygiidae). Animal Behaviour ,.
T,N.W.(.) (). The Natural History of Inbreeding and Outbreeding. Chicago University Press,
Chicago.
T,R.(). Rape in Panorpa scorpionflies and a general rape hypothesis. Animal Behaviour ,.
T,R.(). The relative importance of intra- and inter-specific competition in scorpionfly mating
systems. American Naturalist ,.
T,P.W.(). Courtship disruption modifies mate choice in a lek-breeding bird. Science,Washington ,
.
T,P.W.&A,E.S.(). Active mate choice at cock-of-the-rock leks : tactics of sampling and
comparison. Behavioural Ecology and Sociobiology ,.
T, J. F., G,D.T.&M,G.K.(). Reliable acoustic cues for female mate preference in
a katydid (Scudderia curvicauda, Orthoptera: Tettigoniidae). Behavioural Ecology ,.
Variation in mate choice: a review 
T,G.F.&B,M.T.(). A model of sympatric speciation by sexual selection. Proceedings of the
Royal Society of London,Series B ,.
V, A. C. J., A
$, I., B,A.&R,G.(). Pipefish and seahorses: are they all sex-
role reversed? Trends in Ecology and Evolution ,.
W,R.H.(). Mate guarding by monogamous female razorbills. Animal Behaviour ,.
W, W. E., J., M, A.-M. & C,W.(). Phenotypic variation in the mating preferences of female
field crickets Gryllus integer.Animal Behaviour ,.
W, R. R., W, F., L,P.&  B,E.(). Dynamics of female choice for parental
care in a fish species where care is facultative. Behavioural Ecology ,.
W, W. B., C,P.A.&D,K.(). Female choice of ‘ good genotypes’ as mates is promoted by
an insect mating system. Science,Washington ,.
W, D. M., G,T.C.&W,R.G.(). A product of discriminative learning may lead to
female preferences for elaborate male traits. Evolution ,.
W, C., S, T., B,F.&P,A.J.(). MHC-dependent mate preferences in humans.
Proceedings of the Royal Society of London,Series B ,.
W, L. A., D,P.O.&R,R.J.(). Do female tree swallows guard their mates by
copulating frequently? Animal Behaviour ,.
W, L. A., D,P.O.&R,R.J.(). Testing the mate-guarding hypothesis : a reply.
Animal Behaviour ,.
W,F.&O,I.P.F.(). Lek size, male mating skew and the evolution of lekking. Nature,London
,.
W,R.H.(). Lekking in birds and mammals : behavioural and evolutionary issues. Advances in the Study
of Behaviour ,.
W,R.H.(). Errors, exaggerations and deception in animal communication. In Behavioural Mechanisms
in Evolutionary Ecology (ed. L. A. Real), pp. . University of Chicago Press, Chicago.
W,G.S.&R,P.R.(). Female choice response to artificial selection on an exaggerated male
trait in a stalk-eyed fly. Proceedings of the Royal Society of London,Series B ,.
W,T.&L,R.E.(). Sexual selection and exaggerated male tail length in birds. American
Naturalist ,.
W,J.F.(). Tactics of mate choice. In Mate Choice (ed. P. Bateson), pp. . Cambridge
University Press, Cambridge.
Z,A.(). The cost of honesty (further remarks on the handicap principle). Journal of Theoretical Biology
,.
Z,M.(). Immunology and the evolution of behavior. In Behavioural Mechanisms in Evolutionary Ecology
(ed. L. A. Real), pp. . University of Chicago Press, Chicago.
Z, M., J, K., T,R.&L,J.D.(). Mechanisms of female choice in red jungle fowl.
Evolution ,.
Z, M., L,J.D.&T,R.(). Effects of experimental manipulation of male secondary sexual
characters on female mating preferences in red jungle fowl. Animal Behaviour ,.
... In addition to being important selective environments for sperm, female sperm storage organs likely vary among individuals, following several lines of evidence. First, since genetic variation is a pre-requisite for evolution, the fact that sperm storage organ morphology evolves suggests that it varies (Jennions & Petrie, 1997). ...
... Simultaneous assessment strategies can give different results from other assessment strategies (Janetos, 1980;Jennions & Petrie, 1997;Muniz & Machado, 2018), and to the best of our knowledge, continuous variation in female preferences has not been modelled with simultaneous assessment with a reasonable (for an internally fertilizing species) number of copulation partners; see Millan et al. (2020) for relevant work with a different assessment model, and Van Doorn et al. (2001) and van Doorn and Weissing (2002) for models relevant for broadcast spawners with high mate sampling. Further work is thus needed to understand how variation in female sperm storage organs impacts selection on sperm. ...
... Previous models have highlighted a broad among-female distribution in the preferred male phenotype as a key element in generating disruptive selection on male traits, as one step that can lead to sympatric speciation (Higashi et al., 1999;Van Doorn et al., 2001;van Doorn & Weissing, 2002;Weissing et al., 2011). Our results suggest that disruptive selection will occur under more limited circumstances than was previously appreciated, since females generally are expected to be somewhat limited in the number of males they can sample (Jennions & Petrie, 1997). We thus support Servedio and Boughman's (2017) assertion that novel insights may be obtained in the sympatric speciation literature by further exploring closedended preference functions and limited female searches, similar to what we have simulated here. ...
Article
Sperm cells are exceptionally morphologically diverse across taxa. However, morphology can be quite uniform within species, particularly for species where females copulate with many males per reproductive bout. Strong sexual selection in these promiscuous species is widely hypothesized to reduce intraspecific sperm variation. Conversely, we hypothesize that intraspecific sperm size variation may be maintained by high among‐female variation in the size of sperm storage organs, assuming that paternity success improves when sperm are compatible in size with the sperm storage organ. We use individual‐based simulations and an analytical model to evaluate how selection on sperm size depends on promiscuity level and variation in sperm storage organ size (hereafter, female preference variation). Simulations of high promiscuity (10 mates per female) showed stabilizing selection on sperm when female preference variation was low, and disruptive selection when female preference variation was high, consistent with the analytical model results. With low promiscuity (2–3 mates per female), selection on sperm was stabilizing for all levels of female preference variation in the simulations, contrasting with the analytical model. Promiscuity level, or mate sampling, thus has a strong impact on the selection resulting from female preferences. Furthermore, when promiscuity is low, disruptive selection on male traits will occur under much more limited circumstances (i.e. only with higher among‐female variation) than many previous models suggest. Variation in female sperm storage organs likely has strong implications for intraspecific sperm variation in highly promiscuous species, but likely does not explain differences in intraspecific sperm variation for less promiscuous taxa. We modelled populations where among‐female variation in sperm storage organ (SSO) size (red) was relatively high or low, compared to among‐male variation in sperm size (blue). The form of selection on sperm depended on the interaction between among‐female SSO variation and female promiscuity level (i.e. number of sperm competitors).
... A female's choice of mate can determine her fitness by influencing the number and quality of her offspring (Bateson 1983;Rosenthal 2017). Female mate choice is a key component of sexual selection as it leads to differential mating success among males and can drive the evolution of exaggerated sexual signals (Andersson 1994;Jennions and Petrie 1997), alternative mating tactics (Alonzo and Warner 2000;Luttbeg 2004), and novel traits and trait loss (Basolo 1990;Tinghitella and Zuk 2009). Female mate choice also drives speciation by generating reproductive isolation between populations (Williams and Mendelson 2010). ...
... Examining the effects of female mating status on mate choice in these other contexts would therefore be a fruitful avenue for future research Second, virgin and mated females may not have differed as predicted by theory because of other factors that influence mate choice that were not included in our study. For example, theory predicts that virgin females should mate indiscriminately to avoid the risk of dying unmated before "trading up" (Jennions and Petrie 1997). Their ability or propensity to do so may, however, depend on factors such as the density and quality of potential mates in the population, how long females can store sperm, and the extrinsic risk of mortality. ...
Article
Full-text available
Studies of female mate choice commonly use virgin females as test subjects, either to control for the effects of mating or because virgin females are presumed to be more responsive to mating cues. Theory predicts that virgin females will be less choosy because they risk dying without mating. Moreover, in many species, females spend more of their lives mated than as virgins. Thus, the exclusive use of virgin females in studies of female mate choice may underestimate the strength or direction of female choice and fail to reflect natural mating decisions. We conducted a systematic meta-analysis of female mate choice studies focusing on three scenarios in which female choice might differ in virgin and mated females: reproductive isolation, inbreeding avoidance, and sexually transmitted disease. Using only virgin females was common (53% of 303 studies). In addition, 38% of studies lacked information on female mating history. Contrary to predictions, we found no evidence that virgin females were less choosy than mated females. Nevertheless, excluding mated females from studies of female mate choice leaves an important gap in our understanding of the role of female preferences in evolution. We therefore encourage future studies of female mate choice to consider the natural context of mate choice and include mated females as test subjects when relevant.
... Sexual ornaments and signals attract mates by providing information about quality. Several models exist explaining the linkage between signals and choice, both from a mechanical (how did this individual come to choose that one?) (Jennions and Petrie 1997) and an evolutionary (why is it better to choose that individual?) perspective (Fisher 1930, Zahavi 1975, Hamilton and Zuk 1982, Andersson and Iwasa 1994. The models agree that signal strength correlates with the evolutionary benefits of mating with the signal producer, and therefore stronger signals are better at attracting mates Iwasa 1994, Andersson andSimmons 2006). ...
... While stronger signals improve mating rates in most, if not all, sexually reproducing taxa (Andersson and Iwasa 1994, Andersson and Simmons 2006, Clutton-Brock 2009, the strength of a signal is relative, not absolute. How one individual is perceived depends on how it compares to its competitors (Jennions and Petrie 1997, Bateson and Healy 2005, Gasparini et al. 2013. This, and previous work showing the importance of location to mate choice (Apollonio et al. 1989, Gibson et al. 1991, Rintamäki et al. 2001, suggests that there may be a hitherto unacknowledged selection pressure involved in the evolution of mate choice: the pressure to be able to identify and move to optimal locations. ...
Article
Full-text available
While the strength of sexual signals is important in mate attraction, the ability to accurately compare signals may also have a major effect on mate choice. Large distances between competitors may reduce competition, as accurate comparison of signals becomes harder. This may be advantageous to weak signallers and detrimental to stronger signallers. We create a mathematical model examining optimal distance from stronger competitors for sexual signallers and test its predictions using the common glow‐worm Lampyris noctiluca. Female glow‐worms are flightless and attract males by glowing. Males prefer the brightest female if two females are close to each other. Our model gave different predictions depending on whether searchers fly or not. The model and experiment showed that weak signallers should move away from competitors and strong signallers should move closer to weaker competitors when searchers fly. In contrast, the model predicted that the distance between competing signallers has no effect when searchers do not fly. This reveals an unexpected spatial competition between strong and weak signallers. We conclude that, while signal strength is important in sexual selection, location in relation to others is similarly important as ornamentation in determining the result of mate attraction.
... We encourage researchers to continue to explore the influences of these contextual variables, but across the gamut of fitness-relevant cues rather than on just one cue at a time (e.g., DeBruine et al., 2010;Sugiyama, 2004a). We also encourage researchers to think broadly about ecological, social, and conditional (i.e., dependent on the perceiver's phenotype) variables that may influence the fitness value of a given cue (see Jennions & Petrie, 1997). For example, despite the productivity of research on shifts in the perception of attractiveness as a function of shifts in mating context, there is a comparative paucity of research investigating how individual differences in mating strategy may predict variability in perceptions of attractiveness. ...
Chapter
The Oxford Handbook of Human Mating covers the contributions and up-to-date theories and empirical evidence from scientists regarding human mating strategies. The scientific studies of human mating have only recently risen, revealing fresh discoveries about mate attraction, mate choice, marital satisfaction, and other topics. Darwin’s sexual selection theory primarily guides most of the research in the scientific study of mating strategies. Indeed, research on the complexities of human mate competition and mate choice has centred around Darwin’s classic book. This book discusses theories of human mating; mate selection and mate attraction; mate competition; sexual conflict in mating; human pair bonding; the endocrinology of mating; and mating in the modern world.
... It is more complicated when the complexity of the actions of mate choice is considered. Recent studies showed that in different contexts, females might use different signs, they may vary in the attention they pay to different signs or in the number of signs they evaluate, and that interaction between signs may obscure preferences for single signs (Kodric- Brown and Nicoletto 2001, Møller et al.1998, Jennions and Petrie 1997. ...
Article
Full-text available
In some mammalian species, chemical cues are essential in communications among conspecific individuals. Mate choice is facilitated by sexual pheromones, which distinguish and select mates' qualities. However, female mate choice mechanisms still need to be clearly understood. This study evaluated the use of olfactory cues in the mating decision in albino mice Mus musculus. Bedding-based olfactory preference test was used for testing female response toward bedding odorants of preferred and non-preferred males. Female visiting rate and duration of the two odorants of select and non-preferred male bedding were measured. The results showed a significant female preference for bedding scents of preferred males, which correlated with female visiting rate and duration in the live test of mate choice. The results suggest that female mice can evaluate potential mate quality and use olfactory cues in mate choice as females discriminate the scents of preferred male beddings from non-preferred ones.
... Similarly, only receptive females of the African cichlid fish Burton's mouth-brooder (Astatotilapia burtoni) preferentially associate with territorial males (Clement et al. 2005). Females should therefore be tested at the same stage to ensure not only receptivity, but also because preference function and the degree of female choosiness (Jennions and Petrie 1997) may vary with the stage of the reproductive cycle resulting from physiological regulations or adaptive plasticity of female choice. ...
Article
Full-text available
Female choice has been documented in many animal taxa, and how we test it has been refined through years of studies on the topic. However, when designing mate choice experiments some variables, surprisingly, often remain overlooked, including receptivity and reproductive stage. Here, we aimed to assess whether the female reproductive stage influences strength and direction of mate choice in the zebrafish, Danio rerio. Females were offered a choice between two males differing in body size. We found that female choice in our experimental setup was significantly repeatable and that females preferred larger males. Nonetheless, the level of choosiness of females was affected by the time since the last spawning. Females spent more time choosing when tested 7 and 10 days after spawning rather than 4 days, indicating a higher receptivity to males from one week after the last spawning. Moreover, females preferred larger males only when tested 7 and 10 days after spawning. Our results suggest that female mate choice should take female receptivity into account, by standardizing time since the last spawning across females. More broadly, this suggests that 7–10 days since the previous spawning is the ideal time interval for zebrafish female receptivity to peak, with implications for facilities and researchers to increase egg production in natural spawning events and manual egg collection. Significance statement The role of pre-copulatory female mate preference has long been recognized in sexual selection. Nonetheless, female receptivity often remains overlooked in mate choice experiments especially in external fertilizing species. In the present study, we investigated if the female reproductive stage affects the strength and direction of female mate choice in an external fertilizing fish, the zebrafish, Danio rerio. We found that, when tested 7 and 10 days after spawning rather than 4 days, females spent more time choosing, demonstrating an increased receptivity to males from 1 week following the last spawning. Furthermore, only at 7 and 10 days after spawning females exibith a clear preference for the bigger males. Our study highlights the importance of considering the female receptivity in future studies assessing mate choices in this and other externally fertilizing species, and also for zebrafish facilities to increase egg production in natural spawning events and manual egg collection.
... Polyandrous strategies, for instance, increase the genetic diversity or quality of offspring (Tregenza & Wedell, 1998;Spencer et al., 1999;Fisher et al., 2006Fisher et al., , 2013. Multiple mating may increase the number or survivability of a females' offspring in variable environments (Sale et al., 2013;Thonhauser, Raveh & Penn, 2014), reduce the chance of fertilization by a poor-quality male (Tregenza & Wendell, 1998), increase the chance of producing offspring that are more attractive to mates and more genetically viable (Jennions & Petrie, 1997, 2000 or guarantee fertilization in species that have low volumes of ejaculate or low numbers of spermatozoa (Kraaijveld-Smit et al., 2002). ...
Article
Polyandry is a commonly utilized strategy in mammalian reproductive systems, where females engage in mating with multiple males to increase either the genetic diversity or quality of their offspring. Some species also exhibit male semelparity, a peculiar life-history trait where all or most males within a population begin to die following mating. The northern quoll (Dasyurus hallucatus) possesses both of these reproductive adaptations, and females of this species may produce up to eight young in a single litter. Here we aimed to quantify whether there was any difference in female quolls' choice of male sires between two genetically distinct populations located in the Pilbara region of Western Australia (one mainland-dwelling and one island-dwelling population). The total parentage exclusion probability was 0.9999 for excluding a candidate parent from parentage, given the genotype of a known mother. Overall, we found that every litter had young resulting from multiple males. In some litters, a different male fathered every offspring. Thus, northern quolls demonstrated a greater level of polyandry than has previously been detected in marsupials. Furthermore, females from the less genetically diverse island population exercised mate choice and preferentially bred with males that were, on average , 20% smaller than a different male randomly sampled from the island. There was no detectable difference with regard to male sire body size in the mainland population, indicating a difference of selective pressures for the two populations.
... The perception of risk may also factor into why receivers are more likely to reject all signalers in tests with more available options (Dougherty and Shuker 2015). Individuals that perceive they are unlikely to have another opportunity to mate later may be more likely to accept an unattractive or otherwise inappropriate mate (Berglund 1994;Jennions and Petrie 1997). Our observation that receivers were less likely to respond phonotactically when there were higher numbers of stimuli in the local environment could be explained by receivers' perception that available prospective mates are plentiful (Shelly and Bailey 1992). ...
Article
Animal communication mediates social interactions with important fitness consequences for individuals. Receivers use signals to detect and discriminate among potential mates. Extensive research effort has focused on how receiver behavior imposes selection on signalers and signals. However, animals communicate in socially and physically complex environments with important biotic and abiotic features that are often excluded from controlled laboratory experiments, including noise. “Noise” is any factor that prevents signal detection and discrimination. The noise caused by aggregates of acoustic signalers is a well-known impediment to receivers, but how many individual signalers are required to produce the emergent effects of chorus noise on receiver behavior? In Teleogryllus oceanicus, the Australian field cricket, we assayed female preferences for a temporal property of male advertisement signals, the number of long chirp pulses, using two-, four-, six-, and eight-choice phonotaxis experiments. We found that, as the number of individual signalers increased, receivers became less likely to respond phonotactically and less likely to express their well-documented preference for more long chirp pulses. We found that very few individual signalers can create a sufficiently noisy environment, due either to acoustic interference or choice overload, to substantially impair female preference expression. Our results suggest that receivers may not always be able to express their well-documented mating preferences in nature.
... However, the strong concentration of males in spawning ponds may offer fewer opportunities for individuals of either sex to choose mates. Thus, sexual selection theory assumes that intense competition should lessen mate choices (Briggs, 2008;reviewed by Lu et al., 2009;Jennions & Petrie, 2007) and weaken non-random mating (Crespi, 1989;Aronsen et al., 2013). Our results showed size-dependent pairing occurred when OSR was slightly male-biased, suggesting that large males may overcome small males more easily endure such attempts from the relatively few smaller males which resulted in more successful pairings by large males. ...
Article
In many animal species, an increase in the operational sex ratio (OSR), density or a combination of both should lead to more intensive competition among individuals of the more abundant sex. To test this, we examined pairing patterns of Minshan’s toad (Bufo minshanicus) from six populations between 2008 and 2015 along the eastern Tibetan Plateau in south-west China. OSRs in breeding aggregations of Minshan’s toad are normally male biased and males actively compete with each other for acquisition and retention of mates. We found evidence that deviations from random mating by size varied between populations and between years according to the magnitude of the OSR and male density. Larger males were generally more successful in pairing than smaller males when the OSR was slightly male biased and male density was high. However, the resulting size-disproportionate mating was more evident when OSR was closer to 1.99, indicating a positive correlation with the intensity of aggressive scramble competition. Thus, the intensity of male-male competition may partly explain variation in size-disproportionate mating among populations.
Article
Costly heterospecific mating interactions, such as hybridization, select for prezygotic reproductive isolation. One of the potential traits responding to the selection arising from maladaptive hybridization is habitat preference, whose divergence results in interspecific habitat segregation. Theoretical studies have so far assumed that habitat preference is a sexually shared trait. However, male and female habitat preferences can experience different selection pressures. Here, by combining analytical and simulation approaches, we theoretically examine the evolution of sex-specific habitat preferences. Habitat segregation can have demographic consequences, potentially generating eco-evolutionary dynamics. We thus explicitly consider demography in the simulation model. We also vary the degrees of species discrimination to examine how mate choice influences the evolution of habitat preferences. Results show that both sexes can reduce hybridisation risk by settling in the habitats where abundant conspecific mates reside. However, when females can discriminate species, excess conspecific male aggregation intensifies male-male competition for mating opportunities, posing an obstacle to conspecific aggregation. Meanwhile, conspecific female aggregation attracts conspecific males, by offering the mating opportunity. Therefore, under effective species discrimination, females play a leading role in initiating habitat use divergence. Simulations typically result in either the coexistence with established habitat segregation or the extinction of one of the species. The former result is especially likely when the species differ to some extent in habitat preferences upon secondary contact. Our results disentangle the selection pressures acting on male and female habitat preferences, deepening our understanding of the evolutionary process of habitat segregation due to hybridization.
Article
In many species, preferences are acquired by a learning process known as sexual imprinting. Recent findings suggest that this process may lead to a preference for a partner that is slightly novel. We propose that an asymmetrical preference for novel partners might provide a mechanism driving evolutionary change in the characteristics used in mate choice
Article
The mating preferences of female seaweed flies Coelopa frigida were determined by observing their acceptance or rejection of males of known size. The inversion karyotype of both males and females was also determined. Females exhibited a preference to mate with large males, and evidence is presented that a genetic correlation exists between the female preference and the preferred trait. Females carrying the inversion karyotype associated with large male size showed a strong preference for large males; females carrying the inversion associated with small male size also exhibited a preference for large males, but it was significantly less strong. This finding suggests that a Fisherian process may be operating.
Article
We studied sexual selection in Lawes' Parotia, a lek-mating bird of paradise, during 1981-1983 in Papua New Guinea. There was a high variance in mating success among males, with fewer than half of the individuals mating in any one year. This variance was independent of male-male interactions and disruptions. A role of female choice in sexual selection was suggested by the patterns of female visitation to courts and statistical correlations across males between phenotypic traits and mating success. Females repeatedly visited most males in their home ranges and began visiting males up to six weeks before mating. In one or more years, six aspects of male behavior and one morphological variable were positively correlated with mating success, but the probability values were not significant using a simultaneous inference test. Calculation of combined probability values across all three years revealed that one aspect of male display behavior, the probability of display, positively and significantly influenced mating status. The probability of display was also significantly correlated with relative mating success among males. Females showed strong fidelity to mates, both within and between seasons. Display sites of male Lawes' Parotia are variably dispersed, but mating success did not differ for grouped and solitary males. These data confirm an important role of female choice in sexual selection in birds of paradise but also suggest that female choice may be unrelated to the process of lek-initiation in this species.