PerspectiveEVOLUTIONARY GENETICS

The Causes of Haldane's Rule

See allHide authors and affiliations

Science  30 Oct 1998:
Vol. 282, Issue 5390, pp. 889-891
DOI: 10.1126/science.282.5390.889

Speciation HN1, HN2, HN3, HN4, the splitting of single evolutionary lineages into reproductively isolated groups, remains one of the most elusive phenomena in evolution HN5, HN6. The process requires thousands to millions of years, and we are confronted with a vast array of demonstrable contributors—geographical isolation, natural selection, sexual selection, and changes in karyotype, to name several—whose relative importance is difficult to untangle (1). The last 10 years have brought noticeable progress in the genetics of speciation—progress that has come because evolutionary geneticists have focused largely on one aspect of speciation: the production of inviable and sterile hybrids and Haldane's rule HN7, HN8. The most tantalizing regularity in animal speciation, this rule derives from Haldane's (2) observation that “When in the F1 [first generation] offspring of two different animal races one sex is absent, rare, or sterile, that sex is the heterozygous [heterogametic; XY, XO, or ZW] sex HN9.” Haldane's rule is a way station through which almost all pairs of animal species pass on their way to producing completely inviable or sterile hybrids (1). In a study on page 952 of this issue, Presgraves and Orr (3) provide near-definitive data that address the causes of Haldane's rule. Contrary to the classic Popperian formula HN10, HN11, HN12 in which science marches forward over the corpses of rejected hypotheses, their report provides empirical support for both of the most widely accepted explanations.

Haldane's rule holds for 99% of 223 cases of sex-specific hybrid sterility and 90% of 115 cases of sex-specific hybrid inviability (3, 4) (see table). If stretched to include cases in which hybrid females and males differ only quantitatively in fertility or viability, the rule extends to even more animals (4) and some dioecious plants (5) HN13. Its generality suggests that a common evolutionary force may underlie speciation in many different groups. The fact that Haldane's rule predicts inviable or sterile (heterogametic) males in Diptera and mammals, but inviable or sterile (heterogametic) females in Lepidoptera and birds, implies a critical role for the sex chromosomes in this intermediate step in speciation, rather than just for gender per se.

View this table:

For more than 40 years, most evolutionists accepted an X chromosome-based explanation of Haldane's rule proposed by Muller (6). Muller's hypothesis built on Dobzhansky's (7) HN14 insight that genetic changes beneficial or harmless in one genetic background may be deleterious in another genetic background, because of negative interactions that had not been screened by natural selection. Thus, over time—on the order of 105 to 106 years—isolated populations accumulate increasing numbers of genetic changes that may be advantageous or neutral within species but that produce sterility or inviability in hybrids. Muller recognized that deleterious interactions between X-linked loci and autosomal loci would affect the sexes differently. In particular, he argued that X-linked incompatibilities, if partially recessive, would more seriously afflict the sex with only one X, as hybrids of that sex would fully display the deleterious recessive effects. This explanation for Haldane's rule—recently elaborated and dubbed the “dominance theory” (8, 9)—held sway until an influential 1985 experiment (10) reawakened interest in Haldane's rule.

Coyne (10) studied two Drosophila HN15, HN16, HN17 hybridizations that obey Haldane's rule for sterility. He created “unbalanced” hybrid females—females who inherited both of their X's from their mother (by using attached-X stocks in which females' X chromosomes are connected and therefore transmitted together to their daughters). He argued that the sterility of hybrid males should imply sterility of the unbalanced hybrid females under Muller's theory, because these females' X-autosome incompatibilities are just as severe as those of hybrid males. Yet unbalanced females were fertile in both tests. This led to a temporary eclipse of the dominance theory and a scramble for alternatives (11).

Several hypotheses were proposed (4, 12), but only one has found much support: the “faster male” theory of Wu and Davis (13). This theory, which attempts to explain only Haldane's rule for sterility in male heterogametic species, is based on a fundamental distinction between the genetics of sterility and the genetics of inviability. Coyne's test implicitly assumed that incompatibilities causing hybrid male and female sterility are identical or at least accumulate at similar rates. Yet we know from work in Drosophila that, although most lethal mutations within species kill both sexes, most sterile mutations are sex-specific. Thus, we might expect different sets of loci to produce male and female hybrid sterility. These loci might well be subject to sex-specific evolutionary pressures and thus may evolve at different rates. Males in particular are often subject to intense sexual selection that contributes to rapid divergence of male phenotypes ranging from plumage to genitalia, possibly accelerating the rate of evolution at loci that produce male-specific sterility. These observations motivate the faster-male hypothesis—that Haldane's rule for sterility in male-heterogametic species follows from the faster accumulation of male-sterilizing than of female-sterilizing incompatibilities. This mechanism is not a universal explanation for Haldane's rule, as it explains neither Haldane's rule for inviability (assuming that the same incompatibilities afflict both sexes) nor Haldane's rule for sterility in female-heterogametic species (in such species, the males still show accelerated evolution for sexual traits, but hybrid females are differentially sterile or inviable). Nevertheless, the faster-male hypothesis gained strong experimental support from two recent genetic introgression experiments on pairs of Drosophila that obey Haldane's rule for sterility (14). Both experiments found that hybrid genotypes created by inserting small segments of the D. mauritiana genome into D. simulans were far more likely to be male-sterile than female-sterile (or inviable in either sex).

Although persuasive, the introgression results involved only one small clade. These experiments are too labor-intensive to replicate broadly, and they do not critically evaluate the relative contributions to Haldane's rule of faster-male evolution and dominance. To test the generality of both theories, Presgraves and Orr (3) analyzed published data from 174 mosquito hybridizations, 34 from the genus Aedes and 140 from the genus Anopheles. In Aedes, a small region of the sex chromosomes determines gender, with the rest of the X and Y chromosomes being homologous and fully functional; thus, almost all genes on the sex-determining chromosomes show autosome-like patterns of inheritance. The second group of mosquitoes, Anopheles, has typical X-Y sex determination, with a genetically inert Y.

Presgraves and Orr's (3) central insight is that faster-male evolution can produce Haldane's rule for sterility in male-heterogametic species, whether or not the males have “typical” hemizygous sex chromosomes. As predicted by the faster-male theory, Aedes hybridizations clearly follow Haldane's rule for sterility, with sterile males appearing in all 11 cases of sex-specific hybrid sterility. Without a large region of the X chromosome being effectively hemizygous, Muller's dominance mechanism should be ineffective in Aedes. As predicted by the dominance theory, Aedes does not display Haldane's rule for inviability. In contrast, in Anopheles, where both the dominance and faster-male mechanisms should act, 21 of the 24 examples of sex-limited hybrid inviability follow Haldane's rule. Anopheles also obey Haldane's rule for sterility in all 56 cases of sex-specific hybrid sterility. Moreover, a larger overall fraction of Anopheles than of Aedes hybridizations follow Haldane's rule, as expected given that in Anopheles two Haldane's rule-producing forces are acting together. Thus, Presgraves and Orr have found strong evidence supporting both of the leading explanations for Haldane's rule.

Our laboratory has provided complementary support for the dominance theory in a recent study of published data from 125 Drosophila hybridizations (15). Of the 125 species pairs, 81 have “small” X chromosomes (including about 20% of the genome) and 44 have “large” X chromosomes (about 40% of the genome). We compared the association between hybrid viability/fertility and “genetic distances,” which crudely estimate evolutionary divergence times, in large-X and small-X pairs. If hybrid dysfunction involves many incompatibilities among loci scattered throughout the genome, large-X pairs should experience more X-autosome problems than small-X pairs for a given amount of genetic divergence. Thus, the dominance theory predicts that hybrid males from large-X pairs will be less fit than hybrid males from small-X pairs, all else being equal. As predicted, Haldane's rule occurs at smaller average genetic distances between large-X pairs than between small-X pairs.

These new results should bolster the conclusion that both faster-male evolution and dominance contribute to Haldane's rule. Of these explanations, however, only the dominance theory can explain Haldane's rule for inviability in any group or Haldane's rule for sterility in female-heterogametic taxa. The dominance theory also provides a natural explanation for two additional phenomena: The repeated finding that the X chromosome appears to contribute disproportionately to hybrid inviability and sterility (the “large X effect”) (11) and the absence of Haldane's rule for sterility in mammals [see (9) for an explanation]. Conversely, faster-male evolution unquestionably plays a role in Haldane's rule for sterility in Drosophila and mosquitoes.

What remains to be done? We still know too little about the genes that cause postzygotic isolation. We lack, for instance, direct genetic analyses assessing whether X-linked incompatibilities in hybrids behave recessively, although several lines of evidence suggest that they do (4, 9, 12, 15). The demonstrated power of analyzing Haldane's rule in diverse taxa should encourage researchers to look more carefully beyond Diptera. New insights may emerge from dioecious plants, such as Silene, with heteromorphic sex chromosomes (5). We also need more genetic analyses of Haldane's rule in female-heterogametic species. If the faster-male mechanism is working in these taxa, as suggested by the pervasive phenotypic evidence for sexual selection in birds and butterflies, it should oppose the appearance of Haldane's rule for sterility. In female-heterogametic taxa, therefore, we might expect to see a relative excess of hybrids producing Haldane's rule for inviability rather than sterility. In fact, the most recent data (3, 4) show that 69% of the 84 examples of Haldane's rule in birds and Lepidoptera involve inviability, in contrast to only 19% of 217 examples in Diptera. Birds and Lepidoptera pose an additional challenge because their sex chromosomes are generally much smaller than those of Diptera. In these groups, how can dominance give rise to Haldane's rule when there are so few X-linked loci (9)? Although the main causes of Haldane's rule now seem clear, many ancillary questions await further analyses.

HyperNotes Related Resources on the World Wide Web

General Hypernotes

The World Wide Web Virtual Library: Biosciences points to virtual library pages for Evolution and Genetics. Each of these pages presents a long list of Web resources.

The MIT Biology Hypertextbook, developed by the Experimental Study Group at the Massachusetts Institute of Technology, provides background information on the biology of cells and genetics. Mendelian Genetics, a section of the hypertextbook, includes a discussion of sex-linked traits.

MendelWeb provides annotated lists of Internet resources for genetics.

The Evolution Revolution is a site designed to educate users about evolutionary theory. A guided tour presents information in a linear format for those people who have little or no previous knowledge of the subject of evolution. The library provides articles about evolution, biographies of evolutionists, timelines of evolutionary events, and a glossary of terms.

Taxonomy and Nomenclature, developed by BIOSIS, is an Internet resource for classification of organisms.

The Origin of Species by Charles Darwin is available on the Web.

Numbered Hypernotes

1. Michael Turelli's Web page describes his research and lists selected publications.

2. The Species Concept, maintained by BIOSIS, is an introduction to the species concept and speciation.

3. An introduction to speciation is presented by Phillip E. McClean

4. Models of Speciation, Genetics of Speciation, and Cases of Speciation are three lectures developed by David Rand for Biology 48: Evolutionary Biology, a course offered at Brown University.

5. Bomis: The Evolution Ring is an annotated list of selected Web sites for the study of evolution.

6. Enter Evolution: Theory and History is a Web-based exhibit on evolution organized by the University of California Museum of Paleontology. The exhibit features biographies of influential researchers in evolutionary biology.

7. “Haldane's rule” by A. H. Orr (in the 1997 Annual Review of Ecology and Systematics) is a review of the preferential sterility or inviability of hybrids of the heterogametic sex.

8. H. Allen Orr's Web page at the University of Rochester describes his research on the genetic causes of Haldane's rule.

9. The Chromosomal Basis of Inheritance by Earl Fleck describes the role of sex chromosomes in determining sex.

10. “Popperian ideas on progress and rationality in science” by John Watkins (in The Critical Rationalist, vol. 2, no. 2, 5 June 1997) is a detailed discussion of Popperian philosophy.

11. A brief biography of Karl Popper is provided by the Department of Philosophy of Ohio State University.

12. The Karl Popper Web provides a brief biographical sketch and links to related information.

13. Mutational Meltdown of Endangered Species describes monoecy and dioecy, heterozygous and homozygous loci, mutation, natural selection, and other phenomena in population genetics. These pages are presented by The Computational Science Education Project.

14. A brief biography of Theodosius Dobzhansky is presented by the MSU EMuseum's Anthropology Biographies.

15. FlyBase is a database of the Drosophila genome.

16. The Berkeley Fly Database includes genome data from the Berkeley and European Drosophila Genome Projects, linked to curated information on genes and aberrations in FlyBase.

17. The Drosophila Virtual Library provides a list of Internet resources for the study of Drosophila. An introduction to Drosophila and its importance in biology is included.

18. Center for Population Biology, University of California, Davis

References and Notes

View Abstract

Stay Connected to Science

Navigate This Article