MARIALUZ CALDERON | Change Password | Change User Info | CiteTrack Alerts | Subscription Help | Sign Out
Summary of this Article
dEbates: Submit a response to this
article
Similar articles found in:
SCIENCE Online
ISI Web of Science
PubMed
PubMed Citation
This Article has been cited by:
 
other online articles
Search Medline for articles by:
West, S. A. || Sheldon, B. C.
Search for citing articles in:
ISI Web of Science (5)
Alert me when:
new articles cite this article
Download to Citation Manager
  
Collections under which this article
appears:
Evolution
EVOLUTION:
The Benefits of Allocating Sex
Stuart A. West, Edward Allen Herre, Ben C. Sheldon*
Evolutionary biologists have developed an excellent
understanding of the selective factors that shape the way in which
a given organism allocates resources to male and female
offspring--a process called sex allocation (1). Studies of sex
allocation have provided explanations for a wide range of
phenomena--for example, the variation among animals in the
proportion of offspring that are male (sex ratio), pollen-ovule
ratios in plants, and the age of sexual transition in organisms such
as certain coral reef fish that change sex during their lifetime (1,
2). The strength of empirical support for the existence of sex
allocation and the selective factors that shape it allows studies of
sex allocation to address more detailed questions about natural
selection. Furthermore, sex allocation theory can be used to
elucidate the population structure and epidemiology of medically
important pathogens such as the protozoan parasites that cause
malaria.
Precision of Adaptation
At a time when school boards in the United States are debating whether to include the theory of
evolution by natural selection as part of the curriculum, studies of sex allocation are providing some of
the best support for this theory (2). There are several reasons for this: (i) sex allocation can often have a
clear, immediate, and direct effect on fitness; (ii) theoretical models are based on relatively simple
trade-offs that often rely on only a small number of key variables; and (iii) the important variables are
usually easy to measure. Moreover, relative to most other traits, sex allocation has the advantage that
predictions of optimal allocation patterns can be derived from first principles that are directly linked to
the most basic elements of evolutionary theory.
For example, extreme sex ratio adjustments in fig-pollinating wasps confirm many of the tenets of
evolutionary theory (3-5). There are many species of fig-pollinating wasps, and in each case, female
Science -- West et al. 290 (5490): 288
file:///C|/Documents and Settings/Administrator/Desk...tions 2000/Science -- West et al_ 290 (5490) 288.htm (1 of 8) [1/25/2002 4:50:57 PM]
wasps pollinate and lay eggs in the enclosed fruit of their own host fig species. Mating occurs between
the wasps that develop in the same fruit, before the females disperse. Typically, if only a single female
lays eggs in a fruit, she produces an extremely female-biased sex ratio (only 5 to 10% of the offspring
are males). As the number of females laying eggs in a fruit increases, the sex ratios in the broods
become less biased (see the figures, immediately below). Although there are deviations between
observed sex ratios and those predicted by theory (6), the fit is often very close.
A fig of one's own. Precision of adaptation and the sex ratios of fig-pollinating wasps. Shown are the
observed sex ratios (circles) and theoretical optima (curved lines) for different numbers of female
fig-pollinating wasps (from three species of the genus Pegoscapus) laying eggs in a fruit. The numbers
next to the circles show the relative frequency with which that number of females lays eggs in a fruit in
nature--that is, in the fig species pollinated by P. herrei, 0.96 of fruit have only one female enter
(pollinate and lay eggs), 0.03 have two females enter, and 0.01 have three enter. The observed sex ratios
of progeny are closest to theoretical situations (number of females laying eggs in a fruit) that are
encountered most frequently (4). The observed shifts in sex ratio are greatest in species where the
number of females laying eggs in a fruit is more variable.
Broody female wasps. The variation (observed/binomial) in sex ratios of offspring when only one
female lays eggs in a fig fruit (the dashed line represents binomial variance). The more frequently only
one female lays eggs in a fig fruit, the less variation there is in the sex ratio of the progeny. [Adapted
from (5)]
Science -- West et al. 290 (5490): 288
file:///C|/Documents and Settings/Administrator/Desk...tions 2000/Science -- West et al_ 290 (5490) 288.htm (2 of 8) [1/25/2002 4:50:57 PM]
The observed deviations from the predicted optimal sex ratio are not random. When different
fig-pollinating wasp species are compared, the mean sex ratio of offspring produced by a given number
of females laying eggs in a fruit is closest to theoretical predictions for the situations (number of females
laying eggs in a fruit) that are encountered most frequently in that species (4). Furthermore, females
show a greater ability to alter their brood sex ratios (in response to variations in the number of females
laying eggs in the same fruit) in those species where the number of females entering a fruit is more
variable (4) (see the figure, below). Finally, females show less variation in the sex ratio of their broods
in the situations (number of females laying eggs in a fruit) that they encounter most frequently (5).
Although these data provide strong evidence for adaptive sex allocation, they also suggest that
organisms vary in the degree to which their behavior fits theoretical optima.
More daughters for inbreeders. (Left) Sex ratios of progeny can be used to estimate the rate of
inbreeding in protozoan blood parasites such as those causing malaria. The predicted sex ratio
(proportion of gametocytes that are male) is plotted against the inbreeding rate (14). When the rate of
inbreeding is high, the sex ratio is constrained by the need to produce enough male gametes to fertilize
female gametes. This constraint is determined by c, the mean number of viable gametes released by a
male gametocyte (c is equivalent to the maximum number of times that a male could mate). (Right) Sex
ratios can be used to estimate the rate of inbreeding in the malaria parasite Plasmodium and in the
intestinal parasite Toxoplasma. Sex ratios of progeny in malaria (and other blood parasites) are
extremely variable, suggesting that the inbreeding rate also varies enormously. This is not unexpected
given that the degree of inbreeding is likely to depend on infection rates: The greater the number of
parasites infecting a host, the lower will be the rate of inbreeding (14). In contrast, the greater female
bias in the sex ratios of the progeny of intestinal parasites such as Toxoplasma suggests higher rates of
inbreeding.
Sex allocation offers excellent opportunities for examining the constraints and limits on adaptation.
Unfortunately, the best understood constraints on adaptation tend to be striking developmental or
phylogenetic examples that are of limited general application. Very little is known about factors limiting
adaptation that could be applied to most organisms (7), such as mutation, antagonistic pleiotropy (genes
that improve one aspect of adaptation while reducing another), and processing of information from the
environment.
The most striking sex ratio patterns have been found in insects, especially the Hymenoptera (ants, bees,
wasps). The haplodiploid genetic system of these insects allows females to control the sex of offspring
by regulating whether eggs are fertilized or not--males are haploid (single set of chromosomes) and
develop from unfertilized eggs, whereas females are diploid (double set of chromosomes) and develop
from fertilized eggs. In contrast, vertebrates rarely exhibit extreme skews in sex ratio, which may reflect
Science -- West et al. 290 (5490): 288
file:///C|/Documents and Settings/Administrator/Desk...tions 2000/Science -- West et al_ 290 (5490) 288.htm (3 of 8) [1/25/2002 4:50:57 PM]
a constraint imposed by chromosomal sex determination (1). However, recent studies of mammals and
particularly birds (for example the Seychelles warbler) have shown some striking shifts in the sex ratio
of offspring (8), suggesting that, contrary to popular assumptions, genetic sex determination is not an
all-powerful constraint on sex allocation.
There are alternative explanations for why fewer cases of extreme sex ratio skews exist in vertebrates. A
fig wasp may be able to assess the number of females currently laying eggs in the same fruit more easily
than a mammal can assess factors influencing sex allocation, such as the amount of lactation that she
will be able to provide or the genetic quality of her mate (9). Extreme shifts in vertebrate sex ratios may
represent cases where variables can be assessed reliably. For example, among Seychelles warblers the
variable is the quality of territory, which is determined by the availability of their food source (insects).
The daughters of warbler offspring help their parents rear subsequent progeny, whereas the sons
disperse. In high-quality territory, having a helper is advantageous and so predominantly daughters are
produced, whereas in low-quality territory the increased competition for food means that a helper is a
disadvantage and so mainly sons are produced. Another complication for vertebrates is that the
combination of factors influencing sex ratio can be complex. This complexity decreases the selective
advantage of shifting the sex ratio in response to any single factor. In most wasps, the selective
consequences of any particular brood sex ratio are immediately realized. By contrast, adult life-spans in
most vertebrates are relatively long, resulting in complications that arise from overlapping generations
(10) or interactions between siblings (6).
Using Sex Allocation to Infer
Characteristics of a Population
Sex allocation provides an easy way to estimate population characteristics that are technically difficult
or expensive to measure directly (11). One example, with potentially important benefits, is the use of
sex allocation patterns in malaria (Plasmodium) and other protozoan parasites to infer the amount of
inbreeding (also called the selfing rate, which is defined as the proportion of a female's daughters that
are fertilized by her sons). Because the rate of inbreeding in these parasitic species can influence the
evolution of resistance to vaccines and drugs, inbreeding estimates are important for designing effective
control and treatment programs (12). Direct measures of the inbreeding rate using molecular genetics
can be difficult to obtain, and past estimates have been extremely controversial (12). Sex allocation
theory allows the inbreeding rate to be estimated more readily because the occurrence of inbreeding
skews the sex ratio of offspring in favor of females, analogous to the situation in fig wasps (13, 14). The
higher the level of inbreeding, the greater is the predicted female bias of the sex ratio. The amount of
inbreeding (F) can be predicted from the observed sex ratio (r) by the refreshingly simple equation F =
1 - 2r. Consequently, if we assume that the sex ratio theory is correct, then the inbreeding rate can be
estimated from sex ratio data. An advantage of this method is that sex ratio data can be collected
relatively easily from a number of populations and species, allowing generalizations to be made (see the
figure, above). In cases where both indirect sex ratio data and direct genetic estimates of the inbreeding
rate are available, they are in quantitative agreement, supporting the use of this approach (14).
More generally, sex allocation theory applied to protozoan parasites provides support for the application
of evolutionary optimization models to infectious disease research (14). Despite having to assume
equilibrium states (not an obvious feature of microparasite populations), simple theory is able to explain
variation in a life-history trait (sex ratio) across a taxonomically diverse range of protozoan parasites
Science -- West et al. 290 (5490): 288
file:///C|/Documents and Settings/Administrator/Desk...tions 2000/Science -- West et al_ 290 (5490) 288.htm (4 of 8) [1/25/2002 4:50:57 PM]
(14, 15). In particular, it provides a clear demonstration of the importance of population structure in
determining natural selection in parasitic protozoans. Theory suggests that the virulence of parasites
should respond to the same changes in population structure (16).
Applying sex allocation to the prediction of characteristics of a population or species has great potential
because inferences can be made about any factor that influences the optimal pattern of sex allocation
(11). For example, it can be very difficult to determine whether animals recognize kin and, if they do,
the cues that are involved. In social insects--ants, bees, and wasps--workers adjust the sex ratios of
offspring in response to their relatedness to the males and females that they are raising, indicating that
workers have a mechanism for accurately recognizing kin (17). To what extent nonsocial insects
recognize kin, and whether kin recognition has facilitated the evolution of social behavior is less well
established. Local mate competition (LMC) theory allows us to test for kin recognition in solitary wasps
and bees: Discriminating females should produce a more female-biased sex ratio if they mate with a
sibling rather than outbreeding with unrelated males (3, 18).
Initially driven by attempts to explain the sex ratios of seemingly obscure insect species, sex allocation
studies are now yielding valuable evidence in support of the theory of evolution by natural selection,
and are also proving important for elucidating the biology of protozoan parasites.
References and Notes
In large populations, each individual of the rarer sex would make a greater genetic contribution to
the next generation (frequency-dependent selection). Hence, equal numbers of males and females
is the evolutionarily stable strategy. However, in many situations a bias in sex allocation is
favored in ways that are clearly predicted by theory (for example, under certain circumstances
some individuals in a population will do better by producing sons, and others by producing
daughters; in other circumstances, all individuals may do better by producing more of one sex)
[E. L. Charnov, The Theory of Sex Allocation (Princeton Univ. Press, Princeton, NJ, 1982); H. C.
J. Godfray and J. H. Werren, Trends Ecol. Evol. 11, 59 (1996); S. A. Frank, Foundations of
Social Evolution (Princeton Univ. Press, Princeton, NJ, 1998) [publisher's information]; D. R.
Campbell, Trends Ecol. Evol. 15, 227 (2000)].
1.  
Studies of sex allocation have provided some of the best quantitative evidence for the relative
importance of selection at the gene, individual, kin, and population levels [J. Seger and J. W.
Stubblefield, in Adaptation, M. R. Rose and G. V. Lauder, Eds. (Academic Press, San Diego, CA,
1996), pp. 93-123 [publisher's information]; W. D. Hamilton, Narrow Roads of Gene Land I,
Evolution of Social Behaviour (Freeman, Oxford, 1996) [publisher's information]; M. Chapuisat
and L. Keller, Heredity 82, 473 (1999)] [Medline].
2.  
W. D. Hamilton, in Reproductive Competition and Sexual Selection in Insects, M. S. Blum and N.
A. Blum, Eds. (Academic Press, New York, 1979), pp. 167-220; E. A. Herre, Science 228, 896
(1985); S. A. Frank, Evolution 39, 949 (1985).
3.  
E. A. Herre, Nature 329, 627 (1987).4.  
S. A. West and E. A. Herre, Evolution 52, 475 (1998).5.  
The observed variations in sex ratios are predicted by LMC theory. If n females each lay equal
numbers of offspring in a fruit, the predicted sex ratio is (n - 1)(2 - s)/n(4 - s), where s is the
proportion of females that mated with a brother (3). Explanations based on selection at the
6.  
Science -- West et al. 290 (5490): 288
file:///C|/Documents and Settings/Administrator/Desk...tions 2000/Science -- West et al_ 290 (5490) 288.htm (5 of 8) [1/25/2002 4:50:57 PM]
individual and kin level suggest that the female-biased sex ratio is favored because it reduces
competition between brothers. If only one female lays eggs in a fruit, then the extent of LMC is at
its greatest (all males competing for mates are brothers), and so she is predicted to produce a sex
ratio of 0, which is interpreted to mean producing just enough sons to mate with her daughters.
As the number of females laying eggs in a fruit increases, the extent of competition for mates
between brothers (LMC) is reduced, and so a less female-biased sex ratio is favored [W. D.
Hamilton, Science 156, 477 (1967) [Medline]; P. D. Taylor, Nature 291, 64 (1981)].
N. H. Barton and M. Turelli, Annu. Rev. Genet. 23, 337 (1989) [Medline].7.  
J. Komdeur et al., Nature 385, 522 (1997); L. E. B. Kruuk et al., Nature 399, 459 (1999)
[Medline]; B. C. Sheldon et al., Nature 402, 874 (1999).
8.  
Variation in environmental predictability can also explain patterns among different parasitic wasp
species. In many species of parasitic wasp, where only one individual can develop per host,
females lay male eggs on small hosts and female eggs on large hosts. This is presumed to be
advantageous because females gain a greater benefit from the resulting increase in body size.
Females are less likely to adjust the sex of their offspring in response to host size when hosts are
not killed and continue to grow after she has laid her egg. In this case the size of a host is a less
reliable cue of resources available for offspring development [B. H. King, Oecologia 78, 420
(1989)].
9.  
J. H. Werren and P. D. Taylor, Am. Nat. 124, 143 (1984); S. A. West and H. C. J. Godfray, J.
Theor. Biol. 186, 213 (1997).
10.  
The most useful areas for using sex allocation in this way will be those where there is sufficient
knowledge of relevant biology to make correct assumptions, and where extreme sex ratio shifts
can be expected.
11.  
M. Tibayrenc, Adv. Parasitol. 36, 47 (1995) [Medline]; R. E. L. Paul and K. P. Day, Parasitol.
Today 14, 197 (1998).
12.  
In several groups of protozoan parasites the life-cycle has a sexual stage (the gametocyte or
gamont). LMC can favor female-biased sex ratios when mating takes place, not at random within
the population, but among a small fraction of the population (14). For example, in blood parasites
such as malaria (Plasmodium), gametocytes in a single blood meal (imbibed by a mosquito)
usually mate, and so mating only takes place between the same parasite genotype. In contrast, in
the intestinal parasite Toxoplasma, mating usually takes place between different parasite
genotypes infecting the same local area of the host gut. LMC theory applied to protozoan
parasites suggests that the predicted sex ratio (r) can be related to the inbreeding rate (F) by the
equation r = (1 - F)/2. (F, the inbreeding coefficient, is the probability that two homologous genes
in two mating gametes are identical by descent.) The sex ratio is predicted to decline from 0.5 for
complete outcrossing (F = 0) to 0 for complete inbreeding (F = 1).
13.  
A. F. Read et al., Proc. R. Soc. London Ser. B 260, 359 (1995) [Medline]; J. Pickering et al.,
Evol. Ecol. Res. 2, 171 (2000); S. A. West, T. G. Smith, A. F. Read, Proc. R. Soc. London Ser. B
267, 257 (2000) [Medline].
14.  
D. Shutler and A. F. Read, Oikos 82, 417 (1998); R. E. L. Paul et al., Science 287, 128 (2000); S.
Reese and A. F. Read, Trends Ecol. Evol. 15, 169 (2000).
15.  
S. A. Frank, Q. Rev. Biol. 71, 37 (1996) [Medline].16.  
Science -- West et al. 290 (5490): 288
file:///C|/Documents and Settings/Administrator/Desk...tions 2000/Science -- West et al_ 290 (5490) 288.htm (6 of 8) [1/25/2002 4:50:57 PM]
J. J. Boomsma and A. Grafen, Evolution 44, 1026 (1990); U. G. Mueller, Science 254, 442
(1991); L. Sundstrom, Nature 367, 266 (1994); J. D. Evans, Proc. Natl. Acad. Sci. U.S.A. 92,
6514 (1995) [Medline]; L. Sundstrom, M. Chapuisat, L. Keller, Science 274, 993 (1996); L.
Sundstrom and J. J. Boomsma, Proc. R. Soc. London Ser B 267, 1439 (2000) [Medline].
17.  
J. M. Greeff, J. Evol. Biol. 9, 855 (1996).18.  
We thank P. Awadalla, A. Griffin, L. Kruuk, A. Read, S. Reese, T. Solomon, and R. West for
useful comments.
19.  
S. A. West is at the Institute of Cell, Animal and Population Biology, University of Edinburgh,
Edinburgh EH9 3JT, UK. E-mail: stu.west@ed.ac.uk E. A. Herre is at the Smithsonian Tropical
Research Institute, Republic of Panama. B. C. Sheldon is in the Department of Zoology, University of
Oxford, Oxford OX1 3PS, UK. 
Summary of this Article
dEbates: Submit a response to this
article
Similar articles found in:
SCIENCE Online
ISI Web of Science
PubMed
PubMed Citation
This Article has been cited by:
Search Medline for articles by:
West, S. A. || Sheldon, B. C.
Search for citing articles in:
ISI Web of Science (5)
Alert me when:
new articles cite this article
Download to Citation Manager
  
Collections under which this article
appears:
Evolution
This article has been cited by other articles:
West, S. A., Flanagan, K. E., Godfray, H. C. J. (2001). Variable host quality, life-history
invariants, and the reproductive strategy of a parasitoid wasp that produces single sex clutches.
Behav. Ecol. 12: 577-583 [Abstract] [Full Text]
l   
Weiblen, G. D. (2001). HOW TO BE A FIG WASP. Annu. Rev. Entomol. 47: 299-330
[Abstract] [Full Text]
l   
Volume 290, Number 5490, Issue of 13 Oct 2000, pp. 288-290.
Copyright ? 2000 by The American Association for the Advancement of Science. All rights reserved.
Science -- West et al. 290 (5490): 288
file:///C|/Documents and Settings/Administrator/Desk...tions 2000/Science -- West et al_ 290 (5490) 288.htm (7 of 8) [1/25/2002 4:50:57 PM]
AAAS Member Only Site Resume Database
Science -- West et al. 290 (5490): 288
file:///C|/Documents and Settings/Administrator/Desk...tions 2000/Science -- West et al_ 290 (5490) 288.htm (8 of 8) [1/25/2002 4:50:57 PM]