Behavioral Ecology Vol. 12 No. 3: 266-268
© 2001 International Society for Behavioral Ecology
Hamilton Symposium |
Wondering about sex: W. D. Hamilton's contribution to explaining nature's masterpiece
ETH Zürich, Experimental Ecology, ETH-Zentrum NW, CH-8092 Zürich, Switzerland
Address correspondence to P. Schmid-Hempel. E-mail: psh{at}eco.umnw.ethz.ch .
Sexual reproduction, or sex for short, is an extremely successful breeding
strategy. With some exceptions, metazoan organisms use sex, and even among
protozoans or bacteria, some forms of sex exist, defined by its consequence of
gene exchange (Bell, 1982
).
Yet, theory has so far failed to provide a universal and satisfactory
explanation for the adaptive advantages of sex in Darwinian terms. This
question captured William D. Hamilton's imagination for some 20 years,
starting in the mid-1970s. His answer traces back to Haldane
(1949) and was critically
stimulated by George C. Williams
(1975). In Hamilton's view,
short-term advantages resulting from antagonistic coevolution by parasites
favors sex, despite its inherent cost as compared to the asexual
(parthenogenetic) alternative.
An annoying consequence of sex is visible in a well-known myth. When Noah
loaded his ark, he took two individuals of each species—one male and one
female—on board. Given the limited carrying capacity of his ark, Noah
clearly had to accept a twofold cost of sex. Indeed, sexual species are forced
to produce males in addition to females. If males were equally costly to
produce, this requires that half of the resources available for breeding have
to be invested in males whose only function is to fertilize the daughters. In
most species, males provide little parental effort. To make matters worse, and
as a consequence of having males, any one offspring that a female produces
receives only half of her genes, the other half being from her mate. Although
the precise definition often varies between students of the problem, there is
such a thing as a "twofold cost" of sex compared to the asexual
alternative (Maynard Smith,
1978). Bell (1982)
defined sex by its aspects of mixis and syngamy, that is, the merging of
genetic information, generally from different sources, into a single
off-spring. Sex is therefore conceptually different from reproduction because
it changes the genetic state of the cell rather than the number of cells as
happens with reproduction (Bell,
1982). We nevertheless speak of "sexual reproduction"
because in higher organisms these two processes have become inextricably
linked. This is not the end of the dilemma, however. Recombination of genes is
a major defining feature of mixis and syngamy in higher organisms.
Recombination universally breaks up gene combinations and therefore destroys a
successful genotype that has, in fact, managed to survive and is now able to
reproduce. The combination of these adversities makes the adaptive value of
sex hard to explain.
Fisher (1930) realized that
sexual reproduction, in particular the process of genetic recombination, leads
to an increase in the genetic variance among offspring. According to his
fundamental principle of natural selection, the rate of evolution is directly
proportional to the extant amount of (additive) genetic variance present in a
population. Therefore, sexual populations can respond to selection faster than
asexual populations. Another way to put it is that sex and recombination
allows allocating independently arising favorable mutants to one offspring
much more efficiently. Asexuals have to wait for these independent mutations
to occur one after another in the same lineage in order to find them combined
in a single offspring (the "Fisher-Muller model"). This longterm
advantage for a sexual population, so Fisher
(1930) argued, favors sex and
makes it spread and be maintained against the asexual alternative over long
periods of time. This paradigm, which in essence was also shared by August
Weismann and Hermann Muller, remained very much unchallenged for nearly 30
years. However, in the rebellious climate of the 1960s, evolutionary
biologists started to doubt the validity of arguments based on long-term
benefits for entire sets of individuals. Indeed, Crow and Kimura
(1965,
1969) realized that Weismann,
Muller, and Fisher all relied on group selection to explain sexual
reproduction.
Group selection arguments of the kind postulated by Fisher
(1930) were diametrically
opposed to Hamilton's (1964)
concept of kin selection and the implied process of direct, short-term
benefits for alternative genetic information. From his own work on extreme sex
ratios in a variety of insect species [e.g., fig wasps
(Hamilton, 1967)], Hamilton
could see that in small, localized and therefore inbred populations, the
consequences of sex for recombination are more or less eliminated. At the same
time, females in such populations can easily reduce their production of males
and thus avoid some of the costs of sex without compromising efficient
reproduction.
Against this background of his dissatisfaction with existing explanations,
Hamilton was asked to review two books that both appeared in 1975: Michael T.
Ghiselin's The Economy of Nature and the Evolution of Sex
and George C. Williams's Sex and Evolution. While he had some
reservations against Ghiselin's treatment, Hamilton found himself in natural
agreement with William's stance that "sex must be shown to be
advantageous to the individual sexist, not just to population or species as a
whole" (Hamilton, 1975;
175). In particular, Hamilton made the remarkable statement that
it seems to me that [to explain sex] we need environmental fluctuations around a trend line of change. For the source of these we may look to fluctuations and periodicities inherent in our solar system, and also to the possibility of others generated by life itself. The latter line of thought tends to carry us back from the egg of sex to the hen of a multispecies system. (Hamilton, 1975
Here, the kernel of the idea of antagonistic coevolution becomes visible, although, at the time, Hamilton did probably not think of parasites as the major cause.
This seemed to have changed radically over the following years and took
shape in his landmark paper on "Sex vs. Non-sex vs. Parasites"
(Hamilton, 1980). There, the
idea of negative frequency-dependent selection caused by coevolving parasites
is spelled out in mathematical terms and shown to be able to favor sexual over
asexual reproduction, at least under some conditions. In this scenario, rare
host genotypes have an advantage because they offer only a small target to the
generally more rapidly evolving parasites. An implicit requirement is that
parasites and their hosts match up to some degree. In other words, a parasite
type can only infect one or a few host types, while hosts are susceptible to
some but not all parasite types. This matching must reflect some genotypic
variation in the host (and parasite, for that matter) because genotypic
variation is what is affected by sex. In any case, when rare host genotypes
have an advantage, rare host types must increase in numbers. At some point,
this inevitably leads to the loss of their advantage due to rarity, and the
parasites will now have ample opportunities to catch up on this common host
and increase in numbers. Some time later still, the formerly rare host
genotypes have become heavily infested by their coevolving parasites and will
be at a disadvantage. These hosts will therefore start to decline in numbers,
until they have reached their former status of a rare but fit genotype. In the
meantime, other host (and parasite) genotypes have also gone through this
co-evolutionary cycle. The time lag between the change in the host frequencies
and the capacity of parasites to respond by numbers causes host and parasite
genotypes to track each other over time. This scenario can create persistent
cycles, albeit of irregular shape, with characteristics depending on the exact
conditions. More importantly for our discussion, mothers that produce their
offspring sexually are more likely to produce, by recombination, rare
genotypes for their offspring than asexual mothers that have to wait for
mutations to do the same. Therefore, sexual mothers are more likely to have
offspring that escape the currently prevailing parasite types—the
immediate advantage for the individual sexist that Hamilton was looking for
(Hamilton, 1993;
Hamilton et al., 1990).
It often happens that, at certain times during the history of a science,
new ideas are somehow in the air. This was the case for the problem of sex in
the mid-1970s. For example, Levin
(1975), considering pest
pressures on plants, proposed that recombination, preventing the congealing of
the genome into a single linkage group, was selected for by persistent
tracking of plant hosts by multiple pathogens and herbivores. A decisive
element in the discussion was added by Clarke
(1976) and Jaenike
(1978) by pointing out that
recombination is probably not advantageous simply because it produces new
genotypes in offspring but because it generates rare genotypes. This is the
essential idea of negative frequency-dependent selection whereby the rare
genotypes have a high fitness and the common ones a low fitness. It was
Hamilton who fleshed out these ideas in the way we discuss them today.
Van Valen (1973) realized
that the geometric distribution of life spans of species, genera, and families
over geological time spans, as inferred from palaeontological records, defied
any simple notion of how accumulating effects of some kind (i.e., some form of
"senescence") could lead to the ultimate death of a species.
Rather, such a time-independent risk of extinction could be much more
convincingly explained by assuming an ongoing coevolutionary arms race between
a species and its competitors and enemies. This is very much like Alice's
attempts to follow the Red Queen in Through the Looking Glass by
running as fast as she can just to discover that both still are at the same
place. When Van Valen (1973)
used this analogy he did not think so much of parasites in this context. Bell
(1982) connected this term to
the explanation of sex and especially referred to the temporal dynamics of
coevolving hosts and parasites, in contrast to the spatial aspect of
among-offspring competition (which he called the "tangled bank").
It is interesting that the implications of the original Red Queen metaphor of
Van Valen (1973) and the
concept of Bell (1982) are
actually quite different. In the coevolutionary race envisaged by Van Valen,
species evolve in some direction—for example, toward harder shells in
mussels and bigger claws in crabs. The essential feature of host-parasite
coevolution, however, is the reuse of genetic information without any apparent
evolutionary direction (Hamilton et al.,
1990). Therefore, viewed from the outside, species may not appear
to evolve at all, while behind this Potemkinian facade there is a violent
turnover and recycling of genes as parasites chase their hosts through the
genotype space.
Hamilton developed his ideas further in the early 1980s. He used a
combination of analytical treatment and computer evaluations to consider
explicit models for the evolution of sex
(Hamilton, 1980). Essentially
similar conclusions were also derived in a later study
(Hamilton et al., 1981). In
these studies, a major problem had to be discussed, too. At the time, models
showed that the best conditions for the spread of sex were found when
parasites exert strong, truncating selection and hosts have high fecundity.
However, these are not the most obvious correlates of sex in nature. Indeed,
sexual species typically have low fecundities; that is, they are species of
large body size and extensive parental care (such as humans), and most
parasites do not kill but rather just debilitate the host. However, the
analysis in Hamilton et al.
(1981), and especially later
in Hamilton et al. (1990)
showed that such conditions are not prohibitive for sex to prevail.
Hamilton was deeply interested in a special property of the Red Queen
scenario that can explain the maintenance of large amounts of genetic
variation in natural populations by selection rather than neutrality. In fact,
compared to rivaling hypotheses, such as the mutation-accumulation hypothesis
(Kondrashov, 1982), the Red
Queen-type coevolutionary scenario suggests that sexual populations stow away
temporarily unfit genetic information for a while because such alleles are not
eliminated but protected by negative frequency-dependent selection. These
alleles necessarily become rare with time but can provide protection during
the next, though occasional episode where the selective environment reverses
its state (i.e., new types of parasites become common). Hence, whether sex
spreads is affected more often by which genotypes occupy the lower end of the
fitness scale rather than who occupies the higher end. This is a consequence
of the fact that the long-term geometric mean fitness determines the fate of a
sexual or asexual variant and not the arithmetic mean fitness. And here, sex
fares better than asex, because, after a while, individuals in sexual
populations can still generate a rare offspring genotype when the overall
parasite pressure on common types has become strong, whereas the asexual
parents carrying the now needed alleles were eliminated (and can only be
regenerated by the vastly slower process of mutation). Hence, despite a higher
fitness that asexual variants can exploit by carrying the best alleles most of
time, they are unlikely to persist through occasional crunch periods when
severe parasite pressure against these successful types has built up. During
these episodes, recombination furnishes the now advantageous combinations much
more quickly than mutation does. As a side effect, Hamilton realized, genetic
variation is maintained in the population.
Empirical support for the concept of a parasite-driven Red Queen process is
difficult to gather, especially in field systems. The most convincing evidence
so far comes from the New Zealand freshwater snail Potamopyrgus
antipodarium, where a long-term study has provided evidence for a cost of
sex (Jokela et al., 1997a),
but also for the connection of sex with parasites (especially trematodes;
Lively, 1989) and
rare-genotype advantage (Lively and
Dybdahl, 2000). At the same time, plausible alternative hypotheses
could be eliminated (Jokela et al.,
1997b; Lively et al.,
1998). Nevertheless, Hamilton's theory for the evolution and
maintenance of sex is not universally accepted and, in fact, is a matter of
heavy dispute (e.g., Barton and
Charlesworth, 1998). On the other hand, his vision assembles a
number of disparate phenomena under one umbrella—for example, the
combinatorial lock-and-key aspects of host defenses against parasites and the
advantage of sex through recombination. The field is thus wide open to
imaginative research.
Hamilton's vision extended beyond the simple consideration of the
conditions for the evolution of sex. In fact, Hamilton formulated one of the
most challenging statements during the Dahlem conference in Berlin in 1982,
when he stated that "if the idea about parasites is right, species may
be seen in essence as guilds of genotypes committed to free fair exchange of
biochemical technology for parasite exclusion"
(Hamilton, 1982: 271). How
coevolution with parasites may promote speciation and how this process could
maintain species boundaries remains a major challenge for the future (e.g.,
Breeuwer and Werren, 1995).
Similarly, and perhaps more disturbingly, for behavioral ecologists, Hamilton
et al. (1981: 363) found that
"if sex is so important then our reliance on coefficients of relatedness
in genetical kinship theory is placed in doubt: the coefficients of
relatedness currently used fail to asses special advantages possessed by
sexual progeny." Indeed, the special combinatorial (epistatic)
properties generated by recombination may not be adequately captured by the
average genetic relatedness between parents and offspring. If such epistatic
effects are strong, as they might be during occasional periods of intensive
selection by parasites, the nonlinearity in the selection profile generated by
coevolving parasites can counteract kinship benefits. To cooperate with close
kin is therefore both a boon and a bane
(Baer and Schmid-Hempel,
1999).
There is little doubt that Hamilton's interest in the evolution of sex was a straightforward extension of his thinking on kin selection and on a range of other phenomena, such as skewed sex ratios, group formation, and migration. His unifying principle was that selection operates primarily on genes and over short time scales. While he always stressed that selection operates at any level and all the time, this principle gives selection for the benefit of groups, populations, or species much less weight most of the time. The application of this principle has made behavioral ecology a very successful branch of research. However, we should remind ourselves that Hamilton's legacy is much broader than a series of single concepts that address, for example, the evolution of sociality or sexual selection. Rather, thinking in populations, with their ecology and dynamics of genes, based on sound natural history, is at the heart of the matter to explain the adaptive value of behaviors, or, more generally, the adaptive value of decisions made by organisms in their environment. Whatever direction the field of behavioral ecology takes in the future, this essential distillate of Bill Hamilton's ideas will be with us for a long time to come.
ACKNOWLEDGEMENTS
I am grateful to Boris Baer, Curt Lively, and Jukka Jokela for discussions on Red Queens. My personal contacts with Bill Hamilton will always be fondly remembered.
REFERENCES
Baer B, Schmid-Hempel P, 1999. Experimental variation in polyandry affects parasite loads and fitness in a bumblebee. Nature 397: 151-154.
Barton NH, Charlesworth B, 1998. Why sex and
recombination? Science 281:
1986-1990.
Bell G, 1982. The masterpiece of nature. Berkeley: University of California Press.
Breeuwer JAJ, Werren JH, 1995. Hybrid breakdown between two haplodiploid species: the role of nuclear and cytoplasmic genes. Evolution 49: 705-717.
Clarke BC, 1976. The ecological genetics of host-parasite relationships. In: Genetic aspects of host-parasite interactions (Taylor AER, Muller R, eds). Oxford: Blackwell; 87-103.
Crow JF, Kimura M, 1965. Evolution in sexual and asexual populations. Am Nat 99: 439-450.
Crow JF, Kimura M, 1969. Evolution of sexual and asexual populations. A reply. Am Nat 193: 89-91.
Fisher RA, 1930. The genetical theory of natural selection. New York: Dover.
Haldane JBS, 1949. Disease and evolution. Ricer Sci 19 (suppl): 68-76.
Hamilton WD, 1964. The genetical evolution of social behavior, I and II. J Theor Biol 7: 1-16; 17-32.[ISI][Medline]
Hamilton WD, 1967. Extraordinary sex ratios.
Science 156:
477-488.
Hamilton WD, 1975. Gamblers since life began: barnacles, aphids, elms. Q Rev Biol 50: 175-180.
Hamilton WD, 1980. Sex vs. non-sex vs. parasite. Oikos 35: 282-290.[ISI]
Hamilton WD, 1982. Pathogens as causes of genetic diversity in their host populations. In: Population biology of infectious diseases (Anderson RD, May RM, eds). Berlin: Springer; 269-296.
Hamilton WD, 1993. Haploid dynamic polymorphism in a
host with matching parasites: effects of mutation/subdivision, linkage, and
patterns of selection. J Hered 84:
328-338.
Hamilton WD, Axelrod A, Tanese R, 1990. Sexual
reproduction as an adaptation to resist parasites (a review). Proc Natl
Acad Sci USA 87:
3566-3573.
Hamilton WD, Henderson PA, Moran NA, 1981. Fluctuation of environment and coevolved antagonist polymorphism as factors in the maintenance of sex. In: natural selection and social behavior (Alexander RD, Tinkle DW, eds). New York: Chiron Press; 363-381.
Jaenike J, 1978. An hypothesis to account for the maintenance of sex in populations. Evol Theor. 3: 191-194.
Jokela J, Lively CM, Dybdahl MF, Fox JA, 1997a. Evidence for a cost of sex in the freshwater snail Potamopyrgus antipodarum. Ecology 78: 452-460.
Jokela J, Lively CM, Fox JA, Dybdahl MF, 1997b. Flat reaction norms and "frozen" phenotypic variation in clonal snails (Potamopyrgus antipodarum). Evolution 51: 1120-1129.
Kondrashov AS, 1982. Selection against harmful mutations in large sexual and asexual populations. Gen Res 26: 221-235.
Levin DA, 1975. Pest pressure and recombination systems in plants. Am Nat 109: 437-451.
Lively CM, 1989. Adaptation by a parasitic trematode to local populations of its snail host. Evolution 43: 1663-1671.
Lively CM, Dybdahl MF, 2000. Parasite adaptation to locally common host genotypes. Nature 405: 679-681.[Medline]
Lively CM, Lyons EJ, Peters AD, Jokela J, 1998. Environmental stress and the maintenance of sex in a freshwater snail. Evolution 52: 1482-1486.
Maynard Smith J, 1978. The evolution of sex. Cambridge: Cambridge University Press.
Van Valen L, 1973. A new evolutionary law. Evol Theor 1: 1-30.
Williams CG, 1975. Sex and evolution. Princeton, New Jersey: Princeton University Press.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||