J. theor. Biol. (1975) 53, 205-214
Mate Selection-A Selection for a Handicap
Institute for Nature Conservation Research, Faculty of Life Sciences,
The George S. Wise Centre for Life Sciences, Tel-Aviv University, Israel
(Received 23 July 1974, and in revisedform 2 December 1974)
It is suggested that characters which develop through mate preference
confer handicaps on the selected individuals in their survival. These
handicaps are of use to the selecting sex since they test the quality of the
mate. The size of characters selected in this way serve as marks of quality.
The understanding that a handicap, which tests for quality, can evolve
as a consequence of its advantage to the individual, may provide an
explanation for many puzzling evolutionary problems. Such an inter-
pretation may provide an alternative to other hypotheses which assumed
complicated selective mechanisms, such as group selection or kin selection,
which do not act directly on the individual.
In his theory of sexual selection, Darwin (1874) tried to explain the evolution
of characters such as the antlers of deer, the tail plumes of the peacock, the
brilliant colouration of many birds and their fantastic displays and songs,
by the cumulative effect of females preference for certain male types. He
suggested that the disadvantages to male survival induced by such characters,
are compensated for by more or better females preferring that individual to
other males. But Darwin could not satisfactorily explain why females should
prefer certain males. He just assumed that they prefer certain male types
The theory of sexual selection aroused and still arouses much debate.
There is a basic difficulty to be explained. On the one hand, it is a common
observation that the most beautiful males of a bird species, or the deer
with the largest antlers, are preferred by females, and, on the other hand,
there is no simple explanation to suggest in what ways the preferred males
should be of better quality than others. Wallace (1889), therefore, dismissed
altogether the theory of sexual selection by mate preference while others,
like Poulton (1890), defended it.
Fisher (1930) suggested that initially there was a correlation between the
character preferred by the female and the quality of the male. This correlation
when appreciated by discriminating females can account for the initial
attractiveness of the male. Once sexual selection begins to have its effect,
it generates a rapid chain of events in which the preference itself creates a
selective pressure which accounts for the exaggerated development of the
character and its additional attractiveness. Fisher suggested that (at the
end of the process) males with the exaggerated sexually selected characters
are more attractive to females not necessarily because they are better now
than they were earlier but because they and their sons are more attractive to
females. In Fisher’s words “an additional advantage is conferred by the
female’s preference. The intensity of the second preference will itself be
increased by selection so long as the sons of females excercising the pre-
ference most decidedly have any advantage over sons of other females
whether this is due to the first or the second cause” (my own italics).
Fisher’s argument that males bearing ornaments which might handicap
them are attractive because their sons have a greater chance to be attractive,
contrasts with his own statement that “tastes of organisms like other organs
and faculties must be regarded as the products of evolutionary change
governed by the relative advantage which such tastes may confer”.
O’Donald’s model (1962), in line with Fisher’s argument, assumed that
females evaluate males through a single character. It is obvious that with only
one measurement at hand an exaggeration of the character, beyond its
significance as an indicator for quality, will be undetected by females. But
under most circumstances females probably select males by more than one
character. In fact males advertise to females by voice, colour, movement
and form, etc. An exaggeration of only one of these characters, without
correlation to the quality of the males, should loose its effect by a negative
selection. Any female which continues to prefer such a character will end
up with a worse mate than females which choose males by all other characters
except the exaggerated one. Generally it should be assumed that a character
in a male is attractive to a female because it helps the female to select the
Maynard Smith (1958) already remarked that “sexual selection will have
evolutionary consequences only if those individuals which have character-
istics which make them successful, in the competitions for a mate, are also
fitter than the average as parents”. Williams (1966) compared the process
of mate selection to an “evolutionary battle of the sexes” which “fosters a
skilled salesmanship among the males and an equally well-developed sales
resistance in the females”.
It is difficult to see how Maynard Smith and Williams could agree with
Fisher, that a state of comparative stability may be reached, when the
attractiveness of males may be the main reason why they are preferred
(Fisher’s second cause).
I suggest that sexual selection is effective because it improves the ability
of the selecting sex to detect quality in the selected sex. The selecting sex
benefits because it can be assured of the quality of its mate, while the selected
sex benefits because it can better advertise its quality and thus probably
acquires more or a better mate. But both sexes also lose. Males lose by invest-
ing (time, energy, risks, etc.) in advertising. Females may receive less help
from their mates and bear sons which are less fit to stand the pressure of
natural selection (since they are also of the genotype which invests more in
A survey of a variety of characters which were probably selected by sexual
selection (through mate preference) makes it clear that probably all of them
seem to confer a handicap on survival. The handicap inherent in the effects
of mate preference could also be assumed as a logical conclusion. Before
mate selection achieved its evolutionary effect the organism was in equili-
brium with the pressures of natural selection. If the selective pressure of
mate preference, which has no value to the survival of the individual, is
added to the variety of selective pressures the effect must be negative. The
larger the effect of the preference the more developed the character and the
larger the handicap imposed. Hence a character affected by sexual selection
should be correlated to the handicap it imposes on the individual. The anta-
gonism which exist between the process of natural selection and sexual
selection has already been mentioned by Darwin and Fisher. But they saw the
antagonism as a by-product of the mechanism. My interpretation of sexual
selection implies that sexual selection is effective only by selecting a character
which lowers the survival of the individual. [An argument along this line
was given by Emlen (1973) among several other arguments.] It is possible
to consider the handicap as a kind of a test imposed on the individual. An
individual with a well developed sexually selected character, is an individual
which has survived a test. A female which could discriminate a male possess-
ing a sexually selected character, from one without it, can discriminate
between a male which has passed a test and one which has not been tested.
The more developed the character the more severe was the test. Females
which select males with the most developed characters can be sure that they
have selected from among the best genotypes of the male population
An individual with a very good genotype but without a handicap is
certainly fitter than a handicapped individual which otherwise possesses
the same genotype. But since an individual without the handicapping marker
does not advertise its quality, a potential mate cannot spot it. Females
which choose by a sexually selected character compromise. They select a
good quality male which is handicapped but they can be assured as to
Magnitude of sexually selected character
The effects of mate preference on the evolution of a sexually attractive character.
Quality in the mate is plotted on the vertical axis while the size of the attractive character
(which is considered to be a handicap) is plotted on the horizontal axis. At the intersection
of the axes the character is of the size which fits best the selection pressure of natural
selection. An increase in the character (hence in the handicap) eliminate low quality
individuals (lower hatched area in the diagram). The selecting sex cannot distinguish
quality among potential mates with equally developed attractive character. Note that the
more developed the attractive character the higher the average quality of a potential mate.
their mate’s quality. Sexual selection proceeds in producing an effect, as
long as females benefit more from the assurance of the quality of their
than they lose by mating with a handicapped
and as long as the males
can survive the handicap. It is obvious
males which do
parental care can spend more in order to pass the test of quality (Trivers,
1972; Selander, 1972).
In my model of sexual selection there is no need to assume any special
genetic linkage between the marker of quality and the quality of the individual.
Since whenever the marker is present in a phenotype, the handicap it has
already imposed is a proof that the phenotype and hence very likely also the
genotype is above a certain level of quality marked in the model by the
Since any genetic linkage between a marker and quality is open to a
change and hence to a bluff (i.e. low quality individuals may be tagged by
high quality marker) the test of the handicap is a simple way to exclude
The effects of sexual selection should vary, as suggested already by Darwin,
and have to be “compatible with the existence of the species”. Hence some
species and often one of the sexes (which invest more in parental care) cannot
be easily pushed into a handicap since the pressures of natural selection
are too tight. In these cases mate selection might be achieved either through
an easier test, with a smaller handicap, or without any at all. In the last case
a mate may be selected but “sexual selection” cannot operate. Under such
circumstances mate selection is probably less reliable and takes more time
(Zahavi, in prep., also see below, the comparison of monomorphic and
dimorphic plumage in birds).
Darwin has already pointed out the difficulty in judging in any one par-
ticular case how much of a character developed as a consequence of natural
selection and how much by sexual selection. The effects of both types of
selection are often mixed from the beginning, since sexual selection prefer-
ably acts on characters which have already been correlated with quality.
Natural selection can also affect characters which have evolved by sexual
selection, if these gain some new function. Hence it is almost impossible
to separate clearly the effects of natural selection which are responsible for
the adaptive value of the character from those of sexual selection which
developed its advertising component. A discussion of the effect of sexual
selection will therefore have to be speculative.
The discussion of a sample of characters, which follows, aims to present
some of the possible effects of mate preference. The value of the characters
discussed is explained by the tests they impose on the individuals which
bear them. It is clearly understood that the evolution of all these characters
may bc explained by more than one hypothesis. The discussion of the
examples is given as a model of yet another way to consider the characters.
To determine in any case which of the alternatives is the true explanation,
calls both for a detailed examination of the facts and for further experimental
confirmation. The cases are not presented as a proof of the handicap
principle. My aim is to call the attention to the possibility that the value
of many characters may reside in their action as testing devices.
Characters which attract the other sex cannot always be separated from
those which intimidate rivals of the same sex. Generally the effects of mate
preference should be much larger than those which result from intrasexual
conflicts. The reasons for all this will be discussed elsewhere. Fisher (1930)
came to the same conclusion when he stated that male ornaments acquired
through male-male conflict may be striking but could scarcely ever become
as extravagant as characters which develop to impress a potential mate.
In many species of birds, the female is cryptic while the male is colourful.
The accepted explanation is that females cannot withstand the extra preda-
tion pressure involved in colourful plumage since they have to attend more
to the nest. The colourful plumage of the male is supposed to attract females
and intimidate other males. But there are as many species in which both sexes
are cryptic. Males of such species also mate and fight their rivals. The inter-
esting question is why should some species “find it better” to be excited or
intimidated by the more colourful males? This will only make sense if
colourful males are of better quality. I suggest that a mature, colourful
male has already proved itself to be of a better quality (than one with
cryptic plumage) since it has already withstood the extra predation risk
involved in its plumage. Hence colourful plumage is a mark of quality, and
it is to the advantage of the females to be attracted by colourful males. It is
probably more difficult to discriminate quality among mates with cryptic
plumage. Another way to evaluate the quality of a cryptic male is to observe
it over long periods. Species which have environmental conditions allowing
for prolonged periods of pair formation, or others which cannot put males
to a high predation risk probably, choose such a strategy. This may explain
Lein’s (1973) finding that monomorphic species of warblers take longer
to pair than sexually dimorphic species with colourful male plumage. This
may be the reason why northern duck species, whose short breeding season
demands a short period of pair formation, evolved striking male plumage.
Where time for breeding is short and where males do not participate as
much in parental care, males evolve the colourful plumage as a result of
female preference. The male’s colourful plumage then serves as a mark of
quality and helps females choose good quality males.
Mayr (1972) suggested that the striking male pattern of drakes reduces
interspecies hybridization. Therefore populations, which do not face a
hybridization risk, loose their bright male plumage. But why did not duck
species evolve other isolating mechanisms less risky than bright plumage
patterns ? An alternative hypothesis to that presented by Mayr, is that the
bright male colouration of northern duck species evolved first as an indicator
of the quality of the drake. The colourful plumage, which probably varied
geographically according to various environmental selective pressures,
eliminated hybridization between duck populations to the extent that other
isolating mechanisms were not necessary and hence could not evolve. The
above hypothesis, as to the adaptive significance of the colourful male
plumage, is not presented as conclusive evidence to prove the principle of
testing a mate through a handicap. It is presented to show yet another
way to consider the known facts. The same is also true for the rest of the
examples which follow.
The excessive tail plumes of the peacock which seem to attract the females
obviously deletorious to the survival of the individual. The more brilliant
the plumes the more conspicuous the male to predators, and the longer the
plumes the more difficult it may be for the male to escape predators or to
move about during everyday activity. Hence, only the best males would
be able to sustain the handicap. Therefore, if females select for males handi-
capped by long plumes, they select for quality. It would certainly be better
for females to choose high quality males which were not handicapped by the
plumes. Therefore we have to assume that a discrimination for quality is
more difficult without the test of the plumes.
Many, if not all, sexual displays endanger their performers. Many of them
seem to be designed specifically for that purpose. Warblers sing out of cover.
Lek species dance daily in the open in the same places. The extra risk which
is taken cannot be just to communicate the whereabouts of a male ready
Males probably try to communicate their quality. Since good
quality birds can take larger risks it is not surprising that sexual displays
in many cases evolved to proclaim quality by showing the amount of risk
the bird can take and still survive. If displays had evolved to communicate
in the most efficient way
whereabouts of a bird, in saving energy and
reducing predation hazard, as many alarm caIls have evolved (Marler, 1955),
they would not serve as markers of quality.
Although in this discussion the female is usually referred to as the selecting
sex and the male as the selected sex, it should be born in mind that both
sexes may select and sexual selection may get its effect on both sexes, as is
the case in some monomorphic species. Sometimes the female is more
affected by sexual selection as is the case in quite a number of species
(Selander, 1972; Trivers, 1972).
Coulson (1968) investigated the role of mate preference in a breeding
colony of kittiwakes. These birds fight ferociously with one another to
occupy central territories. Coulson found that birds which breed in the centre
of the colony are of a better quality, as judged by their reproductive per-
formance, than birds breeding in the periphery of the colony. Hence a female
which pairs with a male on a central territory, pairs with a good quality
male. Patterson (1964) found that centrally located territories in the black-
headed gull are more protected from predators so that the fight for central
territories, in that case, could be explained as a consequence of predation
pressure. However, the kittiwakes nest on cliffs and are comparatively free
of predation. Many well-protected sites in the colony are vacant when the
ferocious fight for the central territories is going on. A fight for central sites
in colonies is found also in large sea birds like the gannets which nest on
islands free from predators. An initial tendency to breed together,
in order to get information about food (Ward & Zahavi, 1973), may
give rise to some fights for centrally located sites, since in the centre the
birds are more protected and perhaps get more information. Such fights,
however small, may further evolve through mate preference to serve as an
index of quality. That particular method of testing mate quality has probably
evolved since the ability of colonial birds to fight one another for food is
important. Also, only the better quality birds can depart from offshore
feeding grounds early in the season and succeed in occupying the central
places on the cliffs. Central birds handicap themselves when they invest
time and energy in the fight for the centre but they are compensated by better
Territorial behaviour presents an interesting problem since often males
claim territories which seem to be much larger than the birds may need.
It is not easy to see the selective agent for large territory size. Wynne-Edwards
(1962) suggested that its adaptive significance is to regulate levels of food
supply to population size. He argued that with a large territory the bird
insures the persistence of food supplies to future generations. Such altruistic
behaviour calls for group selection theory. Alternatively, the selection for a
mate may have its effect in this situation. Often males which occupy small
territories seem to be less successful in acquiring a mate (Watson, pers.
comm. ; Zahavi, unpublished observation on Uenunthe). Since a small territory
probably cannot easily support a family, a tendency to select a mate which
occupies a large territory may increase through mate preference to the extent
that males occupy territories much larger than their immediate needs. A fight
for a larger territory must be more difficult than a fight for a small one;
hence, males which succeed in occupying large territories should be of a
better quality than males which occupy small territories. Females which
select such males also select for quality. The conservation of food supply
for the population may be a side effect rather than the causal mechanism
which selects for territory size. O’Donald (1963) already suggested that sexual
selection may be involved in territorial behaviour but he understood some-
what differently the selective mechanism involved.
The contrasting black and white plumage displayed by many adult sea
birds, especially when flying and diving, serves in communication since they
attract more individuals to food sources. The evolution of the bright
“altruistic” plumage and its retention in a population is difficult to explain.
In a predominantly altruistic population, any individual with “selfish” dull
plumage may do better than an “‘altruistic” one, since it will benefit from
sources of food found by all other birds and will not have to share food
found by itself. An indication that mate preference may be involved is
furnished by the fact that only adult birds have the “altruistic” plumage
while immature ones, even when 2 years old, have “selfish” plumage. The
advantage in selecting a mate with “altruistic” plumage is that it selects
an individual which has already passed the test. This test is meaningful
because when breeding in colonies birds have to compete daily over food.
An alternative explanation of the altruistic behaviour may be on the lines
of reciprocal altruism (Trivers, 1971). Such an explanation calls for a mech-
nism of punishments to guard against selfish individuals. But if mate
preference is responsible for the selection of the altruistic character then no
punishment is needed and “altruism” is compensated for by a better quality
It is reasonabie to expect mate preference in a11 sexually reproducing
organisms. As a consequence (if natural selection allows) one would expect
also the evolution of markers for the differentiation of quality in individuals.
O’Donald (1963) suggested that altruistic behaviour may function as a
marker for quality and that it is selected for through mate preference.
Unfortunately that idea has not been followed, perhaps because it was
believed that the evolution of exaggerations (handicaps), through mate
preference, is only a by-product of a “run-away” process (Fisher, 1930;
O’Donald, 1962). When the handicap caused by mate preference is con-
sidered as the key to the selection process rather than its byproduct (Fig. 1)
it is reasonable to expect handicaps, and consequently also altruistic
behaviour, to be widespread phenomena.
Some of the arguments in support of the handicap principle in the evolution
of mate preference are given by Emlen (1973), but at the same time he accepted
the possibility of a random fixation of such characters. He suggested also
that in some cases the importance of the character is in the ability of the
animal to waste its energy to show off its prowess. But according to him
the special means by which it shows its prowess have often been accidentally
The handicap principle as understood here suggests that the marker of
quality should evolve to handicap the selected sex in a character which is
important to the selecting sex, since the selecting sex tests, through the handi-
cap, the quality of its potential mate in characters which are of importance.
Hence the attracting character which evolved through mate preference
should be related to the special ecological problems of the species. The adap-
tive significance of the attracting character should lower the fitness of the
selected sex in relation to the main ecological problems of the species. The
selecting sex should be attracted by a marker only when the handicap it
imposes on its mate and its offspring is smaller than the advantage gained
214 A. ZAHAVI
by securing a better (tested) mate. This evolutionary mechanism need not
be different in its rate from any other selection process. The attracting
characters should evolve only as a consequence of an environmental change
which requires a different kind of test.
I should wish to thank Prof. R. K. Selander for much encouragement and criti-
cism and Prof. Maynard Smith and Dr R. Trivers for discussing the idea. Dr A.
Terkel, Dr Y. Terkel and my wife Dr Avishag Kadman-Zahavi were of much
help both in discussing the idea and in improving its presentation. Thanks are
due to Z. Tamir for typing the manuscript.
COULSON, J. C. (1968).
Man and Selection in Relation to Sex.
2nd edn. London:
EMLEN, J. M. (1973).
Ecology, an Evolutionary Approach.
FISHER, R. A.
(1930). 117ze Genetical Theory
LEIN, M. R. (1973). The biological significance of some communication patterns of wood
warblers (Parulidae). Unnublished Ph.D. Thesis. Harvard.
MAYNARD SMITH, J. (1958). In
A. Barnett, ed.). Cambridge:
Harvard Univ. Press.
MAYR, E. (1972). In
Sexual Selection and the Descent
(B. Campbell, ed.). Chicago :
Aldine Publ. Co.
O’DONALD, P. (1962).
POULTON, E. D. (1890).
Animals: Their Meaning and Use Especially Con-
sidered in the Case
London : Kegan Paul, Trench Trubner & Co.
PATTERSON, I. (1964). Timing and spacing of broods of the black headed gull
L. Ph.D. Thesis, Oxford.
SELANDER, R. K. (1972). In
Sexual Selection and the Descent
(B. Campbell, ed.).
Chicago: Aldine Publ. Co.
R. L. (1971).
Q. Rev. Biol. 46, 35.
TRI~ERS, R. L. (1972). In
Sexual Selection and the Descent
(B. Campbell, ed.).
Chicago: Aldine Publ. Co.
WALLACE, A. R.
(1889). Darwinism: An Exposition
London: MacMillan & Co.
WARD, P. & ZAHAVI, A. (1973).
G. C. (1966). Adagtation and Natural Selection: A Critique
Princeton: Princeton Univ. Press.
Animal Dispersion in Relation to Social Behaviour.
Edinburgh: Oliver & Boyd Ltd.