Sexual Selection, Speciation, and Extinction

Some recent thinking about the conservation of biodiversity emphasizes the
processes that create biodiversity rather than the pattern that happens to be
present today (Mace, Balmford, and Ginsberg 1999). Thus, if we wish to
maintain the capacity to create future biodiversity, we need to understand the
processes responsible. Although the influence of behavior in reproductive isolating
mechanisms has long been recognized, it is only recently that substantial
support has emerged for the importance of sexual selection in sympatric speciation.
Studies of bird speciation show that sexually selected clades (those with
greater sexual dimorphism) are more speciose (Barraclough, Harvey, and Nee
1995; Møller and Cuervo 1998). The cichlid species flocks of the African great
lakes are the classical example of a group that has shown explosive speciation
rates under intense sexual selection (Seehausen 2000).
However, while speciation may proceed rapidly under the influence of
sexual selection, the ornamentation or elaborate displays that are generated
by male intrasexual competition or female mate preferences may predispose
populations that possess them to extinction. Despite the theoretical importance
of sexually selected handicaps (Zahavi 1975, Grafen 1990), empirical
information on costs is accumulating only slowly. However, signaling intensity
and the size of display structures have been shown to have correlated energetic
costs in a number of species (e.g., drumming in wolf spiders [Kotiaho
et al. 1998]) and to affect life history traits (scent-marking frequency is
inversely correlated with growth in mice [Gosling et al. 2000]). The best data

on survival are from experiments on barn swallows showing that survival
prospects of males are inversely related to experimentally manipulated tail
length (Møller 1994). There is also evidence that males carry ornaments at
the expense of their resistance to disease and parasites (Folstad and Karter
1992). Although androgens promote the development of male display structures,
they may also suppress immune function. Experimental evidence for a
trade-off between the sexually selected trait and immunocompetence is now
available in birds, including swallows (Saino and Møller 1996; Saino, Bolzer,
and Møller 1997) and domestic fowls (Verhulst et al. 1999). Whatever the
costs of display traits, all models of sexual selection predict that the evolution
of elaborate display traits involves fitness costs that displace males from their
survival optimum (Møller 2000).
Evidence that sexually selected traits affect extinction rates includes data
supporting Cope’s rule (Cope 1896, Eisenberg 1981), which states that body
size tends to increase within evolutionary lineages and that the risk of extinction
increases with body size. Although Cope’s rule does not apply to all taxa,
it probably has some general application (McLain 1993), and since larger
body size is selected for under intrasexual competition, this effect may be
attributed to sexual selection (Møller 2000). Further evidence comes from the
probability of survival of introduced bird populations: McLain, Boulton, and
Redfearn (1995) found that sexually dichromatic species were significantly less
likely to become established than monochromatic species, perhaps because of
the demographic consequences of the more costly sexually selected display
features. A separate study of introduction success in New Zealand has been
variously explained as a result of the degree of sexual dichromatism (Sorci,
Møller, and Clobert 1998) or of demographic stochasticity, influenced by the
mating system and female choosiness (Legendre et al. 1999).
The loss of biodiversity through an effect on sexually selected traits may
sometimes be inadvertent. For example, the processes of sexual selection that
produced the rich diversity of cichlid fishes in Lake Victoria may be
disrupted by pollution (Seehausen, van Alphen, and Witte 1997). In these
species flocks, reproductive isolation is maintained by mate choice using
colorful signals. When these are obscured in turbid water, interbreeding
between species increases and biodiversity is reduced. Similar arguments
involving natural selection have been made by Endler (1997) about changes
in the light environment of forests with consequent effects on the ability of
cryptically colored animals to escape predation.
Other examples where individual fitness may conflict with population
viability occur in cases of sexual conflict where the outcome may be damaging
for one or both sexes and thus for population growth. Male bean weevils
(Callosobruchus maculatus) damage the genitalia of females during copulation

perhaps to help prevent other males from mating with the same female and,
as a result, female survival is reduced (Crodgington and Siva-Jothy 2000). In
evolutionary arms races between the sexes, an adaptation by one sex that gives
it an advantage (for example in mating) is generally matched by a counteradaptation
by the other sex. However, the outcome of such races can sometimes
favor one sex as revealed in a study of water striders (Heteroptera;
Gerridae) (Arnqvist and Rowe 2002). Male water striders attempt to clasp
females during mating using clasping genitalia, and since there is a cost to
females in being clasped repeatedly after fertilization, females develop counteradaptations
such as abdominal spines. The development of these devices and
corresponding behaviors is generally correlated within species, but detailed
studies of morphology and reproductive behavior show that the advantage for
one sex is greater in some species than in others. This leads to differences in
mating rates and thus potentially to differences in population viability.
Sexual conflict is now recognized as being a central process of evolution
with the potential to shape both speciation and extinction rates (Parker and
Partridge 1998, Arnqvist et al. 2000). Such processes can clearly affect population
viability, but are they accessible to conservation intervention? Direct
intervention to prevent animals from damaging each other is possible in conservation
breeding programs (for example, using advanced reproductive
technology) but can anything be done in the wild? In general, it depends on
the ecological circumstances and whether they can be manipulated. In the
example of polygynous antelopes, there is a potential conflict of interest
between males that aim to mate with as many females as possible and females
that wish to choose between males. Thus females often try to leave territories
and males try to herd them back. The ability of males to monopolize females
in this way depends critically on the distribution of resources: where
resources are concentrated, males can monopolize more females and female
choice is more limited. An example is that of male springbok who defend
territories near water holes in arid areas (Ritter and Bednekoff 1995). Such
behavior potentially leads to reduced effective population size and inbreeding
depression, and could be ameliorated simply by providing more water holes.
Intrasexual competition among females may also have a negative effect on
population viability. For example, in some cavity-nesting ducks, high levels
of brood parasitism may result in lower hatching rates in the population
because of inefficient incubation of very large numbers of eggs and disturbance
by parasitic females. These effects appear to cause declining populations
(Eadie, Sherman, and Semel 1998).
These examples raise the issue of when intervention is ethically acceptable.
It is likely to be less acceptable in species in which a sexually selected
benefit to one sex is threatening extinction, but more likely to be acceptable

when an additional anthropogenic factor is exacerbating the threat. Thus, if
the limited availability of water holes artificially increases the benefit to male
springbok at the cost of inbreeding depression, it may be sensible to manipulate
the distribution of water to ameliorate this effect