Skip to main content

Sexual selection and the evolution of obligatory sex



Among the long-standing conundrums of evolutionary theory, obligatory sex is one of the hardest. Current theory suggests multiple factors that might explain the benefits of sex when compared with complete asexuality, but no satisfactory explanation for the prevalence of obligatory sex in the face of facultative sexual reproduction.

Results and Conclusion

We show that when sexual selection is present obligatory sex can evolve and be maintained even against facultative sex, under common scenarios of deleterious mutations and environmental changes.


Sexual reproduction remains a mystery, and sex as the only mode of reproduction for a species – an even greater mystery. All else being equal, females reproducing asexually have twice the number of their genes transmitted to the next generation, compared with ones that mate with males [1]. Multiple factors, including deleterious mutations [24], adaptation events [57], parasites [8, 9], and combinations of the above [1012] were suggested to explain the benefits of sex against complete asexuality. However, there is no satisfactory explanation for the prevalence of obligatory sex in the face of facultative sexual reproduction. Rather, most of the advantages of sex should still accrue when only a small proportion of the offspring are produced sexually, while the cost of sex would be largely avoided [1316]. The hardest case for the evolution of obligatory sex occurs in the realistic scenario whereby the reproductive isolation between obligatory and facultative sexuals is incomplete. In such a case the long-term advantages of obligaotry sex are shared with the facultative subpopulation, and the maintenance of obligatory sex depends on its short term benefits. No existing model has identified conditions under which obligatory sex would be favored in this scenario. The maintenance of obligatory sex is especially hard to explain when considering mutant alleles which induce sexual reproduction only when the condition of the individual is poor [1719]. Such mutants would pay the cost of sex only when needed, and would have the "abandon-ship" advantage: they would be able to break away from unfit genomes and associate themselves to fitter ones, resulting in a strong short term advantage [18, 19].

Another factor that occurs in sexual populations but not in asexual ones is sexual selection. In many sexual species, the sex that invests less in the offspring (usually the males) has a higher variation in fitness due to more variable mating success [20]. Such sexual selection can be very strong [2123], and has been suggested in the past as a factor related to the evolution and maintenance of sex: Darwin [20] observed the effect of sexual selection in enhancing natural selection. Other authors suggested that differential mating success in males would give females fitter daughters [24], and lead to higher average fitness at equilibrium by reducing mutational load [2528], and improved adaptation to a changing environment [2931].

Sexual selection can thus enable a sexual population to out-compete purely asexual populations in the long run. In this work we explore the possible role of sexual selection in answering the harder question: how obligatory sex could evolve and be maintained when full or partial mixing occurs between obligatory and facultative sexuals.

For an intuition about the short-term effect of sexual selection, let us consider the problematic case of the highly fit females. These females would normally have an immediate twofold fitness advantage when reproducing asexually. When considering only first generation descendents, the benefits of sexual selection are restricted to possibly producing better offspring – a considerable advantage to unfit females, but only a limited gain to highly fit ones. But as early as one generation later, sexual selection may offer females a much greater benefit. Indeed, if males are subject to differential mating success, and if fit females tend to have sexually successful offspring, then the males among a fit female's progeny (sons, grandsons, etc.) may easily produce enough offspring to compensate for the 'loss' she experiences in the first generation (Fig. 1). This advantage of sexual selection first appears in the second generation, and can accumulate over later generations. Of course, the advantage to the fittest females is accompanied by a disadvantage to the least fit ones, whose sons would not be as successful due to the deleterious alleles they inherit. However, a gene for sexual reproduction has a different advantage in these less fit females, stemming directly from genetic mixing – it can dissociate from its bad genetic background and potentially proliferate in future generations [18].

Figure 1

The effect of sexual selection on the progeny of a fit female. A female (represented by a circle) that reproduces asexually ensures that all of her genes are transferred to each of her offspring (A). For a highly fit female that reproduces sexually (B), the proportion of her genes in the offspring is diluted by a factor of two each generation (shrinking black areas). However, if males (represented by squares) have differential mating success, then the amount of progeny fathered by a successful male offspring (bottom right) can more than compensate for the recurring twofold cost.


To study the evolutionary dynamics of a gene coding for obligatory sex, assume we have a locus with two possible alleles. The allele O codes for obligatory sex and the allele F codes for facultative sex. The population is infinite, and individuals have long haploid genomes exposed to deleterious mutations. The population is fully mixed and no inbreeding occurs. Denote the frequency of newborn individuals with i mutations and reproductive strategy X {O, F} by p(i, X).

The number d of new deleterious mutations, added to those inherited by a young individual from its parents, has a Poisson distribution. We neglect mutations in the O/F locus. After mutation we get:

p ˜ ( i , X ) = j = 0 i p ( j , X ) p ( d = i j ) ,  where  d ~ P o i s s o n ( U ) MathType@MTEF@5@5@+=feaagaart1ev2aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacPC6xNi=xI8qiVKYPFjYdHaVhbbf9v8qqaqFr0xc9vqFj0dXdbba91qpepeI8k8fiI+fsY=rqGqVepae9pg0db9vqaiVgFr0xfr=xfr=xc9adbaqaaeGacaGaaiaabeqaaeqabiWaaaGcbaGafmiCaaNbaGaacqGGOaakcqWGPbqAcqGGSaalcqWGybawcqGGPaqkcqGH9aqpdaaeWbqaaiabdchaWjabcIcaOiabdQgaQjabcYcaSiabdIfayjabcMcaPiabdchaWjabcIcaOiabdsgaKjabg2da9iabdMgaPjabgkHiTiabdQgaQjabcMcaPaWcbaGaemOAaOMaeyypa0JaeGimaadabaGaemyAaKganiabggHiLdGccqGGSaalcqqGGaaicqqG3bWDcqqGObaAcqqGLbqzcqqGYbGCcqqGLbqzcqqGGaaicqWGKbazcqGG+bGFcqWGqbaucqWGVbWBcqWGPbqAcqWGZbWCcqWGZbWCcqWGVbWBcqWGUbGBcqGGOaakcqWGvbqvcqGGPaqkaaa@62FB@

The probability ω of survival of a young individual depends only on the number i of mutations it carries: ω i = e-is, where s is the strength of natural selection against each deleterious mutation. We assume that individuals carrying N or more mutations do not survive. For large enough N, the effect of such truncation on the dynamics is negligible. We used N = 50 in our analysis. Control cases with N = 100 yielded indistinguishable results for the parameter range studied.

The frequencies of the different types after mutation and natural selection are

p ( i , X ) = ω i p ˜ ( i , X ) ω ¯ ,  where  ω ¯ = i = 0 N ω i p ˜ ( i , X ) . MathType@MTEF@5@5@+=feaagaart1ev2aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacPC6xNi=xI8qiVKYPFjYdHaVhbbf9v8qqaqFr0xc9vqFj0dXdbba91qpepeI8k8fiI+fsY=rqGqVepae9pg0db9vqaiVgFr0xfr=xfr=xc9adbaqaaeGacaGaaiaabeqaaeqabiWaaaGcbaGaemiCaaNaey4fIOIaeiikaGIaemyAaKMaeiilaWIaemiwaGLaeiykaKIaeyypa0tcfa4aaSaaaeaaiiGacqWFjpWDdaWgaaqaaiabdMgaPbqabaGafmiCaaNbaGaacqGGOaakcqWGPbqAcqGGSaalcqWGybawcqGGPaqkaeaacuWFjpWDgaqeaaaakiabcYcaSiabbccaGiabbEha3jabbIgaOjabbwgaLjabbkhaYjabbwgaLjabbccaGiqb=L8a3zaaraGaeyypa0ZaaabCaeaacqWFjpWDdaWgaaWcbaGaemyAaKgabeaakiqbdchaWzaaiaGaeiikaGIaemyAaKMaeiilaWIaemiwaGLaeiykaKcaleaacqWGPbqAcqGH9aqpcqaIWaamaeaacqWGobGta0GaeyyeIuoakiabc6caUaaa@5EE3@

An individual of type (i, F) invests a of its resources in asexual reproduction, and 1-a in sexual reproduction. An offspring resulting from sexual reproduction has an equal chance of being either a male or a female:

p f e m a l e s ( i , O ) = p m a l e s ( i , O ) = ( 1 C s e x ) p ( i , O ) p f e m a l e s ( i , F ) = p m a l e s ( i , F ) = ( 1 C s e x ) p ( i , F ) ( 1 a ) p a s e x u a l ( i , F ) = p * ( i , F ) a MathType@MTEF@5@5@+=feaagaart1ev2aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacPC6xNi=xI8qiVKYPFjYdHaVhbbf9v8qqaqFr0xc9vqFj0dXdbba91qpepeI8k8fiI+fsY=rqGqVepae9pg0db9vqaiVgFr0xfr=xfr=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@AF32@

Where P females (i, O) is the proportion of individuals in the population which are females of type (i, O), and similarly for the other types and genders.C sex is the cost of sex, assumed to be 0.5 if males contribute nothing but their genes to the offspring.

We assume that due to decreased mating success, males are more severely affected by deleterious mutations than females. The mating probability of a male carrying i mutations is φ i = e-isχ, where χ is the relative strength of sexual selection in comparison with natural selection. Individuals with the O and F alleles are present in the same population and are allowed to interbreed. The frequencies of different male genotypes among the reproducing males, taking into account sexual selection, are

p ^ m a l e s ( i ) = ϕ i / ϕ ¯ ,  where  ϕ ¯ = i = 0 N ϕ i X p m a l e s ( i , X ) . MathType@MTEF@5@5@+=feaagaart1ev2aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacPC6xNi=xI8qiVKYPFjYdHaVhbbf9v8qqaqFr0xc9vqFj0dXdbba91qpepeI8k8fiI+fsY=rqGqVepae9pg0db9vqaiVgFr0xfr=xfr=xc9adbaqaaeGacaGaaiaabeqaaeqabiWaaaGcbaGafmiCaaNbaKaadaWgaaWcbaGaemyBa0MaemyyaeMaemiBaWMaemyzauMaem4CamhabeaakiabcIcaOiabdMgaPjabcMcaPiabg2da9GGaciab=v9aMnaaBaaaleaacqWGPbqAaeqaaOGaei4la8Iaf8x1dyMbaebacqGGSaalcqqGGaaicqqG3bWDcqqGObaAcqqGLbqzcqqGYbGCcqqGLbqzcqqGGaaicuWFvpGzgaqeaiabg2da9maaqahabaGae8x1dy2aaSbaaSqaaiabdMgaPbqabaGcdaaeqbqaaiabdchaWnaaBaaaleaacqWGTbqBcqWGHbqycqWGSbaBcqWGLbqzcqWGZbWCaeqaaOGaeiikaGIaemyAaKMaeiilaWIaemiwaGLaeiykaKcaleaacqWGybawaeqaniabggHiLdaaleaacqWGPbqAcqGH9aqpcqaIWaamaeaacqWGobGta0GaeyyeIuoakiabc6caUaaa@66E7@

Assuming free recombination and a very long genome, the probability of both parents carrying the same mutation is very low. The number of mutations carried by a newborn can thus be approximated by a binomial random variable, taking the success probability to be 0.5 and the number of trials to be the total number of mutations in the two parents. The frequencies of the different types in the next generation are then:

p ' ( i , O ) = j , k = 0 N b ( i | k + j ) [ p f e m a l e s ( j , O ) p ^ m a l e s ( k , O ) + 0.5 p f e m a l e s ( j , O ) p ^ m a l e s ( k , F ) + + 0.5 p f e m a l e s ( j , F ) p ^ m a l e s ( k , O ) ] MathType@MTEF@5@5@+=feaagaart1ev2aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacPC6xNi=xI8qiVKYPFjYdHaVhbbf9v8qqaqFr0xc9vqFj0dXdbba91qpepeI8k8fiI+fsY=rqGqVepae9pg0db9vqaiVgFr0xfr=xfr=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@B1C0@
p ' ( i , F ) = j , k = 0 N b ( i | k + j ) [ p f e m a l e s ( j , F ) p ^ m a l e s ( k , F ) + 0.5 p f e m a l e s ( j , F ) p ^ m a l e s ( k , O ) + + 0.5 p f e m a l e s ( j , O ) p ^ m a l e s ( k , F ) ] + p a s e x u a l ( i , F ) . MathType@MTEF@5@5@+=feaagaart1ev2aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacPC6xNi=xI8qiVKYPFjYdHaVhbbf9v8qqaqFr0xc9vqFj0dXdbba91qpepeI8k8fiI+fsY=rqGqVepae9pg0db9vqaiVgFr0xfr=xfr=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@C396@

The model can be extended to consider the effect of environmental changes. Let us assume that the environment has one of four states and the suitability of an individual to a particular environment is determined by two loci with two alleles each, where one of the four allele combinations in these loci constitutes a perfect match to the environment. The number j of mismatches between the alleles at these loci and the current environment affects both the viability of the individual and the mating probability in the case of males: The viability of an individual carrying i deleterious mutations and j mismatches with the environment would be ω i = e-is-jt, where t is the strength of selection against each mismatch with the environment. The mating success of a male carrying i deleterious mutations and j mismatches would be φ i = e(-is-jt)χ. In this version of the model we study the frequencies p(E, i, X) of individuals with genotype E in the loci affecting environmental match, i deleterious mutations, and reproductive strategy X (see Additional file 1 for the full form of Equation (5) in this case).


Numerically analysing the equations in the case of unchanging environment, we found that sexual selection can favour the short term evolution of obligatory sex on the background of facultative sexuality when genome mutation rate is high enough, even if the cost of sex is significant (Figure 2, different costs of sex and constant environment).

Figure 2

The quantitative effect of sexual selection on fixation of an allele for obligatory sex. Plot of the critical genome mutation rate (U) above which obligatory sex fixes in a facultative population, as a function of the strength of sexual selection (χ). Contour lines with different costs of sex C sex = 0.1,0.2, 0.3, 0.4, and 0.5 are shown. The white area above the C sex = 0.5 contour is the parameter range in which obligatory sex fixes even with a two-fold cost of sex. The plot was generated using a = 0.99, s = 0.05, and N = 50, with free recombination. Fixation of the allele O was defined as reaching a frequency higher than 99.9% within 2500 generations.

The effect was sensitive to the frequency of asexuality, a, among the facultative sexuals. Higher values of asexuality among the facultatives allow obligatory sex to evolve more easily (see Figure 3, different contours representing different values of a). However, our model assumes free recombination, resulting in a dramatically increased effect for facultative sexuals that rarely reproduce sexually. Furthermore, infinite population means that any amount of sexual reproduction is sufficient to produce all possible genotype combinations. The critical values of a in our analysis are therefore likely to greatly overestimate its critical values in finite populations, and in particular in populations where the rate of recombination is limited.

Figure 3

Conditions for fixation of an allele for obligatory sex in populations with different levels of facultative sex. The contours represent the critical mutation rate above which obligatory sex fixes. Contour lines are shown for different frequencies of sexual reproduction among the facultative sexuals: a = 0.1,0.4, 0.7, 0.9, and 0.99. The white area above the a = 0.1 contour is the parameter range in which obligatory sex fixes even in a population where only 0.1 of the offspring are produced asexually. The plot was generated using C sex = 0.2, s = 0.05, and N = 50, with free recombination. Fixation of the allele O was defined as reaching a frequency higher than 99.9% within 2500 generations.

Using the extended model (see Additional file 1) we found that environmental changes can significantly expand the parameter range under which obligatory sex evolves, compared with deleterious mutations acting alone. The effect of environmental changes varies in magnitude, depending both on the selection t on adaptation loci (compare the different contours in Figure 4A) and on the frequency of environmental changes (compare Figure 4A with 4B). When environmental changes occur less frequently, the benefit gained from faster adaptation diminishes, and the advantage to the obligatory sexual individuals is weakened.

Figure 4

Conditions for fixation of an allele for obligatory sex under environmental changes. The contours represent the critical mutation rate above which obligatory sex fixes. Contour lines with t = 0,0.2 and 0.4 are shown. The white area above the t = 0 contour is the parameter range in which obligatory sex fixes even in a constant environment. The plot was generated using C sex = 0.5, a = 0.99, s = 0.05, and N = 50, with free recombination. Environmental changes occurred every 10 generations (A) or every 25 generations (B). Fixation of the allele O was defined as reaching a frequency higher than 99.9% within 2500 generations and retaining it for four consecutive cycles of environmental change.


The advantages of sex that result from sexual selection are inherently different from those resulting from recombination or segregation. The latter two can increase the efficiency of natural selection indirectly, by affecting the distribution of different genotypes in the population [32, 33]. But they do not offer a successful female any benefit from reproducing sexually herself. Sexual selection, on the other hand, results in a direct advantage to the fittest genotypes, if and only if they reproduce sexually.

The first and most direct prediction of this model is that obligatory sex would be more common among organisms with highly differential mating success. In this context, strong sexual selection is not limited to cases of extreme handicaps. It also applies to any case of high choosiness of females or intense competition between males over territory or hierarchy. This prediction is consistent with the general observation that obligatory sex is more common, whereas sexual selection is often stronger, among animals than among plants: Plants do exert intense selection on pollen, which is indicative of the quality of gametes as well as the ability of the mate to produce gametes that are effectively dispersed [34, 35]. Nevertheless, they are inherently limited in their ability to compare the overall quality of potential mates [36]. Second, similar to models for the advantage of sex vs. asexuality, obligatory sex is expected to be more common among organisms experiencing higher genome mutation rates [2, 3]. The genome-wide rate of deleterious mutations in natural populations is still largely controversial [3740]. Recent direct estimates suggest that the rates might be higher than previously expected, and may reach over 1 mutation per genome per generation even in short-lived organisms such as Drosophila melanogaster [41]. Third, our model predicts that obligatory sex would be more common under environmental changes, and in particular under environmental changes with strong effect on the fitness. Parasites may constitute one possible source of such changes [9, 42, 43], and explicit models of obligatory sex under red queen scenarios could provide further insight into their effect.

In our model, male mating success is affected both by deleterious mutations and by environmental changes. While a female cannot easily determine what makes a potential mate maladapted, she often has indications of its overall success in competition over territory, for example [21]. A more complicated situation occurs when male display (e.g. song or colours) does not involve direct competition. However, it has been shown by Lorch et al.[29] that male display is likely to evolve to be condition-dependent, thus affected by both deleterious mutations and environmental changes. Our assumption is therefore that male mating success is correlated with its overall adaptation to its current environment, which does not require direct choice for specific 'good genes'.

One limitation of our model is the assumption of free recombination. Free recombination results in a significantly increased effect of sex, in terms of breaking genetic associations. The effect is especially dramatic for facultative sexuals, particularly ones that reproduce sexually with a low probability. Obligatory sex is therefore likely to be favoured under wider conditions in populations where the level of recombination is limited. Further research using detailed genetic models that allow low levels of recombination and explicit chromosome structure would enable more accurate quantitative analysis.

Our model assumes that overall fitness is positively correlated with mating success. Such associations have been documented in various organisms [4450], but do not apply genome-wide, and there are even conflicts between genes that are advantageous for females and ones that benefit males [51]. However, the correlation between the sexual success of a parent and the success of its offspring does not have to be entirely genetic. For example, it can be mediated through higher maternal investment in the offspring of fitter mothers or more attractive fathers [52, 53]. Under differential maternal investment and male mating success, sex ratio is expected to be male-biased in fitter females [54], as was empirically demonstrated in several cases [55, 56]. We expect obligatory sex to be favoured under wider conditions when sex ratio is thus biased. Finally, the effect of sexual selection on the evolution of obligatory sex is not limited to the specific model of selection presented here. Very similar results are obtained if each female selects the best of n male candidates as a mate.


The fitness benefits of genetic mixing are predicted by many evolutionary models, and some degree of genetic mixing indeed occurs in most organisms, including bacteria and viruses. But for sexual reproduction to become the sole mode of reproduction for so many organisms, an additional factor seems to be required. Sexual selection presents one such factor, offering both short-term and long-term advantages to sexually reproducing individuals. As such, it may have played a key role in the evolution of obligatory sex.


  1. 1.

    Maynard Smith J: The evolution of sex. 1978, Cambridge , Cambridge University Press

    Google Scholar 

  2. 2.

    Kondrashov AS: Deleterious Mutations and the Evolution of Sexual Reproduction. Nature. 1988, 336 (6198): 435-440. 10.1038/336435a0.

    Article  CAS  PubMed  Google Scholar 

  3. 3.

    Muller HJ: The relation of recombination to mutational advance. Mutat Res. 1964, 106: 2-9.

    Article  CAS  PubMed  Google Scholar 

  4. 4.

    Feldman MW, Christiansen FB, Brooks LD: Evolution of recombination in a constant environment. Proc Nat Acad Sci USA. 1980, 77 (8): 4838-4841. 10.1073/pnas.77.8.4838.

    PubMed Central  Article  CAS  PubMed  Google Scholar 

  5. 5.

    Fisher RA: The Genetical Theory of Natural Selection. 1930, Clarendon Press, Oxford

    Google Scholar 

  6. 6.

    Otto SP, Barton NH: The evolution of recombination: Removing the limits to natural selection. Genetics. 1997, 147 (2): 879-906.

    PubMed Central  CAS  PubMed  Google Scholar 

  7. 7.

    Waxman D, Peck JR: Sex and adaptation in a changing environment. Genetics. 1999, 153 (2): 1041-1053.

    PubMed Central  CAS  PubMed  Google Scholar 

  8. 8.

    Hamilton WD, Axelrod R, Tanese R: Sexual reproduction as an adaptation to resist parasites (a review). Proc Nat Acad Sci USA. 1990, 87: 3566-3573. 10.1073/pnas.87.9.3566.

    PubMed Central  Article  CAS  PubMed  Google Scholar 

  9. 9.

    Howard RS, Lively CM: Parasitism, mutation accumulation and the maintenance of sex. Nature. 1994, 367 (6463): 554-557. 10.1038/367554a0.

    Article  CAS  PubMed  Google Scholar 

  10. 10.

    West SA, Lively CM, Read AF: A pluralist approach to sex and recombination. J Evol Biol. 1999, 12 (6): 1003-1012. 10.1046/j.1420-9101.1999.00119.x.

    Article  Google Scholar 

  11. 11.

    Hadany L, Feldman MW: Evolutionary traction: the cost of adaptation and the evolution of sex. J Evol Biol. 2005, 18 (2): 309-314. 10.1111/j.1420-9101.2004.00858.x.

    Article  CAS  PubMed  Google Scholar 

  12. 12.

    Peck JR: A ruby in the rubbish: beneficial mutations, deleterious mutations and the evolution of sex. Genetics. 1994, 137: 597-606.

    PubMed Central  CAS  PubMed  Google Scholar 

  13. 13.

    Charlesworth D, Morgan MT, Charlesworth B: Mutation Accumulation in Finite Outbreeding and Inbreeding Populations. Genet Res. 1993, 61 (1): 39-56.

    Article  Google Scholar 

  14. 14.

    Green RF, Noakes DLG: Is a Little Bit of Sex as Good as a Lot. Journal of Theoretical Biology. 1995, 174 (1): 87-96. 10.1006/jtbi.1995.0081.

    Article  Google Scholar 

  15. 15.

    Hurst LD, Peck JR: Recent advances in understanding of the evolution and maintenance of sex. Trends Ecol Evol. 1996, 11 (2): A46-A52. 10.1016/0169-5347(96)81041-X.

    Article  Google Scholar 

  16. 16.

    Yamauchi A, Kamite Y: Facultative sexual reproduction under frequency-dependent selection on a single locus. J Theor Biol. 2003, 221 (3): 411-424. 10.1006/jtbi.2003.3195.

    Article  PubMed  Google Scholar 

  17. 17.

    Bell G: The masterpiece of nature: the evolution and genetics of sexuality. 1982, Berkeley , University of California Press

    Google Scholar 

  18. 18.

    Hadany L, Beker T: On the evolutionary advantage of fitness-associated recombination. Genetics. 2003, 165 (4): 2167-2179.

    PubMed Central  PubMed  Google Scholar 

  19. 19.

    Hadany L, Otto SP: The Evolution of Condition-Dependent Sex in the Face of High Costs. Genetics. 2007, 176 (3): 1713-1727. 10.1534/genetics.107.074203.

    PubMed Central  Article  PubMed  Google Scholar 

  20. 20.

    Darwin C: The Descent of Man and Selection in Relation to Sex. 1871, New York , D. Appleton

    Google Scholar 

  21. 21.

    Anderson M: Sexual Selection. 1994, Princeton , Princeton University Press

    Google Scholar 

  22. 22.

    Jones JE, Antoniadis E, Shettleworth SJ, Kamil AC: A comparative study of geometric rule learning by nutcrackers: (Nucifraga columbiana), pigeons (Columba livia), and jackdaws (Corvus monedula). Journal of Comparative Psychology. 2002, 116 (4): 350-356. 10.1037/0735-7036.116.4.350.

    Article  PubMed  Google Scholar 

  23. 23.

    Kutsukake N, Nunn CL: Comparative tests of reproductive skew in male primates: the roles of demographic factors and incomplete control. Behavioral Ecology and Sociobiology. 2006, 60 (5): 695-706. 10.1007/s00265-006-0213-1.

    Article  Google Scholar 

  24. 24.

    Trivers RL: Parental investment and sexual selection. Sexual Selection and the Descent of Man. Edited by: Campbell B. 1972, Chicago , Aldine Press, 136–179-

    Google Scholar 

  25. 25.

    Manning JT: Males and the Advantage of Sex. Journal of Theoretical Biology. 1984, 108 (2): 215-220.

    Article  CAS  PubMed  Google Scholar 

  26. 26.

    Atmar W: On the Role of Males. Animal Behaviour. 1991, 41: 195-205. 10.1016/S0003-3472(05)80471-3.

    Article  Google Scholar 

  27. 27.

    Siller S: Sexual selection and the maintenance of sex. Nature. 2001, 411 (6838): 689-692. 10.1038/35079578.

    Article  CAS  PubMed  Google Scholar 

  28. 28.

    Agrawal AF: Sexual selection and the maintenance of sexual reproduction. Nature. 2001, 411 (6838): 692-695. 10.1038/35079590.

    Article  CAS  PubMed  Google Scholar 

  29. 29.

    Lorch PD, Proulx S, Rowe L, Day T: Condition-dependent sexual selection can accelerate adaptation. Evol Ecol Res. 2003, 5 (6): 867-881.

    Google Scholar 

  30. 30.

    Proulx SR: Niche shifts and expansion due to sexual selection. Evol Ecol Res. 2002, 4 (3): 351-369.

    Google Scholar 

  31. 31.

    Whitlock MC: Fixation of new alleles and the extinction of small populations: Drift load, beneficial alleles, and sexual selection. Evolution. 2000, 54 (6): 1855-1861.

    Article  CAS  PubMed  Google Scholar 

  32. 32.

    Kondrashov AS: Classification of Hypotheses on the Advantage of Amphimixis. Journal of Heredity. 1993, 84 (5): 372-387.

    CAS  PubMed  Google Scholar 

  33. 33.

    Barton NH, Charlesworth B: Why sex and recombination?. Science. 1998, 281 (5385): 1986-1990. 10.1126/science.281.5385.1986.

    Article  CAS  PubMed  Google Scholar 

  34. 34.

    Willson MF: Sexual Selection in Plants. The American Naturalist. 1979, 113 (6): 777-790. 10.1086/283437.

    Article  Google Scholar 

  35. 35.

    Snow AA, Spira TP: Pollen Vigor and the Potential for Sexual Selection in Plants. Nature. 1991, 352 (6338): 796-797. 10.1038/352796a0.

    Article  Google Scholar 

  36. 36.

    Bernasconi G, Ashman TL, Birkhead TR, Bishop JDD, Grossniklaus U, Kubli E, Marshall DL, Schmid B, Skogsmyr I, Snook RR, Taylor D, Till-Bottraud I, Ward PI, Zeh DW, Hellriegel B: Evolutionary ecology of the prezygotic stage. Science. 2004, 303 (5660): 971-975. 10.1126/science.1092180.

    Article  CAS  PubMed  Google Scholar 

  37. 37.

    Lynch M: Perspective: spontaneous deleterious mutation. Evolution. 1999, 53: 645-663. 10.2307/2640707.

    Article  Google Scholar 

  38. 38.

    Eyre-Walker A: Evidence of selection on silent site base composition in mammals: Potential implications for the evolution of isochores and junk DNA. Genetics. 1999, 152 (2): 675-683.

    PubMed Central  CAS  PubMed  Google Scholar 

  39. 39.

    Denver DR, Morris K, Lynch M, Thomas WK: High mutation rate and predominance of insertions in the Caenorhabditis elegans nuclear genome. Nature. 2004, 430 (7000): 679-682. 10.1038/nature02697.

    Article  CAS  PubMed  Google Scholar 

  40. 40.

    Gong Y, Woodruff RC, Thompson JN: Deleterious genomic mutation rate for viability in Drosophila melanogaster using concomitant sibling controls. Biology Letters. 2005, 1 (4): 492-495. 10.1098/rsbl.2005.0364.

    PubMed Central  Article  CAS  PubMed  Google Scholar 

  41. 41.

    Haag-Liautard C, Dorris M, Maside X, Macaskill S, Halligan DL, Charlesworth B, Keightley PD: Direct estimation of per nucleotide and genomic deleterious mutation rates in Drosophila. Nature. 2007, 445 (7123): 82-85. 10.1038/nature05388.

    Article  CAS  PubMed  Google Scholar 

  42. 42.

    Hamilton WD: Sex versus non-sex versus parasite. Oikos. 1980, 35: 282-290. 10.2307/3544435.

    Article  Google Scholar 

  43. 43.

    Hamilton WD, Zuk M: Heritable True Fitness and Bright Birds - a Role for Parasites. Science. 1982, 218 (4570): 384-387. 10.1126/science.7123238.

    Article  CAS  PubMed  Google Scholar 

  44. 44.

    Welch AM, Semlitsch RD, Gerhardt HC: Call duration as an indicator of genetic quality in male gray tree frogs. Science. 1998, 280 (5371): 1928-1930. 10.1126/science.280.5371.1928.

    Article  CAS  PubMed  Google Scholar 

  45. 45.

    Petrie M: Improved Growth and Survival of Offspring of Peacocks with More Elaborate Trains. Nature. 1994, 371 (6498): 598-599. 10.1038/371598a0.

    Article  CAS  Google Scholar 

  46. 46.

    Byers JA, Waits L: Good genes sexual selection in nature. Proceedings of the National Academy of Sciences of the United States of America. 2006, 103 (44): 16343-16345. 10.1073/pnas.0608184103.

    PubMed Central  Article  CAS  PubMed  Google Scholar 

  47. 47.

    Evans JP, Kelley JL, Bisazza A, Finazzo E, Pilastro A: Sire attractiveness influences offspring performance in guppies. Proceedings of the Royal Society of London Series B-Biological Sciences. 2004, 271 (1552): 2035-2042. 10.1098/rspb.2004.2815.

    Article  Google Scholar 

  48. 48.

    Drickamer LC, Gowaty PA, Holmes CM: Free female mate choice in house mice affects reproductive success and offspring viability and performance. Animal Behaviour. 2000, 59: 371-378. 10.1006/anbe.1999.1316.

    Article  PubMed  Google Scholar 

  49. 49.

    Bluhm CK, Gowaty PA: Social constraints on female mate preferences in mallards, Anas platyrhynchos, decrease offspring viability and mother productivity. Animal Behaviour. 2004, 68: 977-983. 10.1016/j.anbehav.2004.01.013.

    Article  Google Scholar 

  50. 50.

    Forsman A, Hagman M: Calling is an honest indicator of paternal genetic quality in poison frogs. Evolution. 2006, 60 (10): 2148-2157.

    Article  PubMed  Google Scholar 

  51. 51.

    Chippindale AK, Gibson JR, Rice WR: Negative genetic correlation for adult fitness between sexes reveals ontogenetic conflict in Drosophila. Proc Natl Acad Sci U S A. 2001, 98 (4): 1671-1675. 10.1073/pnas.041378098.

    PubMed Central  Article  CAS  PubMed  Google Scholar 

  52. 52.

    Sheldon BC: Differential allocation: tests, mechanisms and implications. Trends Ecol Evol. 2000, 15 (10): 397-402. 10.1016/S0169-5347(00)01953-4.

    Article  PubMed  Google Scholar 

  53. 53.

    Burley N: The Differential-Allocation Hypothesis - an Experimental Test. American Naturalist. 1988, 132 (5): 611-628. 10.1086/284877.

    Article  Google Scholar 

  54. 54.

    Trivers RL, Willard DE: Natural Selection of Parental Ability to Vary the Sex Ratio of Offspring. Science. 1973, 179 (4068): 90-92. 10.1126/science.179.4068.90.

    Article  CAS  PubMed  Google Scholar 

  55. 55.

    Cameron EZ: Facultative adjustment of mammalian sex ratios in support of the Trivers-Willard hypothesis: evidence for a mechanism. Proceedings of the Royal Society of London Series B-Biological Sciences. 2004, 271 (1549): 1723-1728. 10.1098/rspb.2004.2773.

    Article  Google Scholar 

  56. 56.

    Sheldon BC, West SA: Maternal dominance, maternal condition, and offspring sex ratio in ungulate mammals. American Naturalist. 2004, 163 (1): 40-54. 10.1086/381003.

    Article  PubMed  Google Scholar 

Download references


The authors wish to thank Amir Beker, Gal Chechik, Dan Cohen, Ilan Eshel, Steve Hendrix, Michael Lachmann, John Logsdon, Joanna Masel, Dan Tzur, Joshua Weiner, and three anonymous referees for their helpful comments on drafts of this manuscript. This research was supported in part by NSF grant 0639990.

Author information



Corresponding author

Correspondence to Lilach Hadany.

Additional information

Authors' contributions

Both authors contributed equally to the formulation of the model, the analysis, and the writing of the manuscript.

Electronic supplementary material

Authors’ original submitted files for images

Rights and permissions

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Reprints and Permissions

About this article

Cite this article

Hadany, L., Beker, T. Sexual selection and the evolution of obligatory sex. BMC Evol Biol 7, 245 (2007).

Download citation


  • Sexual Selection
  • Sexual Reproduction
  • Mating Success
  • Deleterious Mutation
  • Sexual Population