Haldane's rule

If, within a species, there is a slight tendency for two lines (varieties, races) to diverge as separate species, then copulation is unimpaired, and healthy children are produced, but some of those children may be sterile (unable themselves to produce children). Haldane noted that when the offspring ("F1 hybrid") of a cross between a male parent from one line and a female parent from the other line is sterile although otherwise healthy, it will tend to be of the heterogametic sex. This is Haldane's rule for hybrid sterility. If some gametes are produce by the F1 hybrid they will be of both types. Thus, among its offspring, albeit reduced in number, the usual 50% : 50% male/female ratio is not disturbed. There is no preferential production of one sex.

If, within a species, the divergence between the two parental lines becomes sufficient to generate genic differences, but copulation is unaffected, then parental gene products may fail to cooperate during development of the fertilized ovum, resulting in hybrid inviability. Remarkably, here often the 50% : 50% male/female ratio is disturbed. There is preferential production of one sex, which is usually the homogametic sex. The heterogametic sex is absent or rare. This is Haldane's rule for hybrid inviability.

If speciation occurs within one geographic region, the first stage of the speciation process is complete when all hybrids, although viable, are fully sterile (hybrid sterility). As species differentiation further progresses, parental genic incompatibilities arise and hybrids fail to be formed. They fail to develop as embryos or to reach adult maturity (hybrid inviability). So the question of their sterility does not arise. There are no mature offspring to be sterile!

As species differentiation further progresses, parental genic differentiations generate anatomical, physiological or psychological differences that result in mating incompatibilities. The "postzygotic barriers" (hybrid sterility and hybrid inviability), are replaced by "prezygotic barriers" (e.g. mating incompatibilities). So the questions of offspring inviability and sterility do not arise. There are no young offspring to be inviable, or mature offspring to be sterile!

Two Schools of Thought Regarding the Form of Hybrid Sterility Associated with Most Speciation Events

CHROMOSOMAL

GENIC

Romanes 1886

Bateson 1904

Kholodkovskii 1912

Darlington 1932

Gulick 1932

Goldschmidt 1940

White 1978

King 1993

Dobzhansky 1934

[MAINSTREAM OPINION]

Coyne & Orr 1998

"In 1912 [Petersburg entomologist Nicolai] Kholodkovskii suggested that the origin of the two biological species of Chermes -- may have been based on their different number of chromosomes which made them incapable of interbreeding -- . Iurii Filipchenko - Dobzhansky's mentor in genetics - supported this idea (Filipchenko 1916)"

Krementsov (1994) Dobzhansky and Russian entomology. The origin of his ideas on species and speciation. The Evolution of Theodosius Dobzhansky (Princeton Univ. Press)

"A New Paradigm"?

"Two of the strongest patterns in evolutionary biology, Haldane's rule and the large effect of the sex chromosomes on postzygotic isolation, still lack wholly convincing explanations. It seems possible that some simple explanation for both phenomena has eluded everyone."

Jerry A. Coyne (1992) Genetics and speciation. Nature 355, 511-515.

"The concept of genic incompatibility as a cause of postzygotic isolation -- forms an integral part of most hypotheses, either implicitly or explicitly."

Cathy C. Laurie (1997) The weaker sex is heterogametic: 75 years of Haldane's rule. Genetics 147, 937-951.

"What makes hybrid male sterility of great current interest is the increasing evidence that the building blocks of this isolating barrier may be radically different from what we had come to believe -- . Most probably, they [the building blocks] are not a more or less abundant collection of genes with individually detectable effects on hybrid fertility (multigenic basis), but a large number of interacting gene sets, made out of minor factors whose individual -- [action] has virtually no effect (polygenic basis). It is clear that a new paradigm is emerging, which will force us to, first, revise many conclusions of past studies that had gathered almost unanimous agreement, and second, try a completely different experimental approach to unveil the genetic basis of hybrid male sterility"

Horacio F. Naveira & Xulio R. Maside (1998) The genetics of hybrid male sterility in Drosophila. In: Enless Forms: Species and Speciation. Oxford University Press. pp. 330-338.

For more on History of Haldane's Rule Click Here

Bateson's Meditations

"Whenever sexual organisms breeding together produce a mixture of forms, there is, -- prima facie reason to suspect that the mixture is due to differentiation of germs. The most familiar case is sex itself. A population consisting of males and females has so many features in common with the differentiating offspring resulting from the segregation of characters among the germ-cells of cross-bred organisms that it is impossible to avoid the suspicion that the two phenomena are similar in causation. A categorical proof of this conclusion would make a remarkable advance in biology."

William Bateson (1906) Application to the Balfour Fund for a research grant . [For earlier thoughts of Bateson (Click Here)]

"We may feel fairly sure that the distinction between the sexes depends on the presence in one or other of them of an unpaired factor -- .The results of genetic research are so bewilderingly novel -- . In all the discussions of the stability and fitness of species who ever contemplated the possibility of a wild species having one of its sexes permanently hybrid [XY or WZ chromosomes]? -- Who would have supposed it possible that the pollen cells of a plant could be all of one type, and it egg cells of two types? Yet Miss Saunders' experiments have provided definite proof that this is the condition of certain [dioecious] stocks [where the female is the heterogametic sex] -- ."

William Bateson (1908) in 'The Methods and Scope of Genetics'. Inaugural Lecture at the University of Cambridge.

"We have now to admit the further conception that between the male and female sides of the same plant these ingredients may be quite differently apportioned, and that the genetical composition of each may be so distinct that the systematist might without extravagance recognize them as distinct specifically [mistake the different sexes as distinct species]. If then our plant may by appropriate treatment be made to give off two distinct forms, why is not that phenomenon a true instance of Darwin's origin of species?"

William Bateson (1922) address in Toronto on 'Evolutionary Faith and Modern Doubts'.

Goldschmidt's Meditations

"If -- genic balance does not account for the facts, we must turn -- to an effect of the chromosome as a whole, an effect of the whole chromosome with its specific serial pattern upon the whole reaction system of the plant. No genic balance is needed and, as for that, no genes either, to understand the reported facts, as soon as we have decided to assume that a chromosome of a definite structural pattern is acting as a unit." (pp. 233)

"Sexual differences within a species may be of such a nature that, if found distributed among different organisms, they would provide a basis for classification into different species, families, or even higher categories. -- Two forms found in nature, which showed morphological differences of such a degree as that existing between male and female insects -- would never be considered as belonging to the same species, or even genus. -- In the sexual differences we have, then, two completely different reaction systems in which the sum total of all the differences is determined by a single genetic differential. -- The genetics of sex determination ought, therefore, to furnish information on how a completely different reaction system may be evolved on the basis of existing -- genetical agencies." (pp. 234)

"It is not this or that gene or array of genes which is acting to produce the extreme morphological differences of the sexes, but rather the typical serial pattern within the X-chromosome, or definite parts thereof. The chromosome as a whole is the agent, controlling whole reaction systems (as opposed to individual traits). The features which are assumed by many geneticists to prevent a scattering of individual sex genes by crossing over (one X in heterogametic sex, or inert Y) actually prevent major changes in the pattern within the chromosome as a whole. -- I must emphasize that such a conception offers mental difficulties to those steeped in the classical theory of the gene." (pp. 236)

Richard Goldschmidt. The Material Basis of Evolution (1940) Yale University Press.

Hybrid sterility, due to failure of meiotic pairing, might reflect initial (C+G)% differences between the regions concerned with sexual differentiation (i.e. differences in base composition would be the "building blocks" referred to by Naveira and Maside, above). This hypothesis differs from the main line of chromosomal hypotheses for hybrid sterility in that the chromosomal differentiations are due to changes in single base pairs (not visible by the light microscope), and not to changes in chromosomal segments as segments per se (sometimes visible by the light microscope).

Such (C+G)% "incompatibilities at the genomic level -- would be manifest at meiosis when parental genomes attempt to recombine, and would affect hybrid fertility", (i.e. sterility, not viability; Forsdyke, 1996; (Click Here). See also:

Forsdyke 1999a Abstract (Click Here);
Forsdyke 1999b. J. Theor. Biol. 201, 47-61 (Click Here);
Forsdyke 2000. J. Theor. Biol. 204, 443-452 (below).

A suitable experimental test of the hypothesis presented in this page, would be progressively to change the base composition [(C+G)%] of a line and examine the appearance of hybrid male sterility.

Perhaps, this test has already been carried out in the form of the "introgression" of chromosomal segments between lines. Thus, analysis of hybrids between two fruit fly lines (D. buzzatii and D. koepferae) produced results which were "quite unexpected".

Let us call the lines A and B, and suppose that the result of a cross between them is the production of just females, not males. Now we want to introduce successively larger pieces of the B genome into the A genome, producing lines A', A'', A''', etc. Then we cross the original line A with A' and see if the ratio of males to females among the offspring is decreased. Then we cross A with A'' and check the ratio again -- and so on. This will allow us to tell how much of the B genome we have to transfer (a) to begin decreasing the ratio, and (b) to eliminate males entirely.

The procedure known as "introgression" allows the experiment to be performed. This involves repeated back-crossing the hybrid offspring (female) of an A x B cross, with males of the parental line A. The haploid gametes of the parental line can be represented as A and A, whereas the haploid gametes of the hybrid offspring would be, not just A and B, but, as a result of recombination during meiosis, A-with-varying-amounts-of-B, and B-with-varying-amounts-of-A.

The gamete type A-with-varying-amounts-of-B united with a parental-line type A gamete ("back-cross") would result in offspring producing gametes of type A-with-even-lesser-amounts-of-B. The process is repeated for many generations to produce the above-mentioned A, A'', A''', etc. stock.

Next, Naveira and Maside assessed the degree of B introgression into A, by taking advantage of the fact that salivary gland cell chromosomes are easily visualized (polytene chromosomes), and the degree of pairing between paternal and maternal homologs can be seen. Where there was B introgression pairing was defective. They noted:

"A correlation between the lengths of individually introgressed segments and the fertility/sterility outcome was observed: long segments produced sterility, while short segments did not -- . On average, interspecific substitution of roughly up to 30% of any of the autosomes allows hybrid male fertility, whereas a 40% substitution leads to hybrid male sterility. --

The transition from fertility to sterility as the length of the introgression is increased can be shown to be quite sharp, thus revealing the existence of genuine threshold length intervals. What is more, the threshold can be crossed by recombining in the genome of a hybrid male two fertile introgressions from distant locations in the same chromosome, or even from different chromosomes."

From further studies with with introgressed non-coding DNA in other lines they further conclude that:

"The coding potential of the introgressions -- might be also irrelevant for hybrid male fertility. It might be only a question of foreign DNA amount -- "

Despite this suggestion, they refer to their model of the "genetic basis of hybrid male sterility in Drosophila" as "generally polygenic," to distinguish it from the "genic" model. They are not inclined on the basis of their evidence to refer to the model as "chromosomal".

"Current evidence suggests that the genetic basis of hybrid male sterility in Drosophila is generally polygenic, and the total number of sterility factors must probably be numbered at least in the hundreds. The individual effect on fertility of any of these factors is virtually undetectable, but it can be accumulated to others. So, hybrid male sterility results from the co-introgression of a minimum number of randomly dispersed factors (polygenic combination). The different factors linked to the X, on the one hand, and to the autosomes, on the other, are interchangeable -- . Recent experiments on the nature of these polygenes suggest that the coding potential of their DNA may be irrelevant."

Sexual Differentiation is on the Path to Species Differentiation:

Haldane's Rule: Hybrid Sterility Affects the Heterogametic Sex First because Sexual Differentiation is on the Path to Species Differentiation

Forsdyke, D. R. (2000) J. Theor. Biol. 204, 443-452

[Copyright permission held by Academic Press. Please see caveats on Home-Page] [Re: quotations, unless otherwise stated italics and words within square brackets are those of the present author, and should not be attributed to the person quoted.]

Introduction

Sex Chromosomes

Similarity between Species and Sexual Differentiations

Haldane's Rule

A Way Station

Chromosomal Hypothesis for Haldane's Rule

Genic versus Chromosomal

General Model

Further Questions

End Note (2005)

End_Note_(Dec_2010)

End_Note_(2016)

End_Note_(Feb._2017)

Summary

Prevention of recombination is needed to preserve both phenotypic differentiation between species and sexual phenotypic differentiation within species. For species differentiation (speciation), isolating barriers preventing recombination may be prezygotic (gamete transfer barriers), or postzygotic (either a developmental barrier resulting in hybrid inviability, or a chromosomal-pairing barrier resulting in hybrid sterility). The sterility barrier is usually the first to appear and, although often initially only manifest in the heterogametic sex (Haldane's rule), is finally manifest in both sexes. For sexual differentiation, the first and only barrier is chromosomal-pairing, and always applies to the heterogametic sex. For regions of sex chromosomes affecting sexual differentiation there must be something analogous to the process generating the hybrid sterility seen when allied species cross.

Explanations for Haldane's rule have generally assumed that the chromosomal-pairing barrier initiating evolutionary divergence into species is due to incompatibilities between gene products ("genic"), or sets of gene products ("polygenic"), rather than between chromosomes per se ("chromosomal"). However, if chromosomal incompatibilities promoting incipient sexual differentiation could also contribute to the process of incipient speciation, then a step towards speciation would have been taken in the heterogametic sex.

Thus, incipient speciation, manifest as hybrid sterility when "varieties" are crossed, would appear at the earliest stage in the heterogametic sex, even in genera with homomorphic sex chromosomes (Haldane's rule for hybrid sterility). In contrast, it has been proposed that Haldane's rule for hybrid inviability needs differences in dosage compensation, so could not apply to genera with homomorphic sex chromosomes.

In many species males and females are produced in equal quantities. Mendel himself noted that such equal quantities would be produced if a recessive homozygote were crossed with a heterozygote (Bateson & Saunders, 1902; Bateson 1909). Thus, if red is dominant to white, when a homozygous white (WW, producing one type of gamete, W and W) is crossed with a heterozygous red (RW, producing two types of gamete, R and W), on average equal numbers of red and white progeny should be produced (RW, WW). If sex were similarly determined, this simple scheme would suggest that one sex be homozygous and the other sex be heterozygous for alleles of a particular gene.

As long as only one allele pair were required there would be no reason to regard this process as different from any other genetic process. The chromosome pair containing the gene, -- let us call them X and Y chromosomes, -- would be equal in all respects, including size ("homomorphic"). One sex would be homozygous for the recessive allele (the "homogametic" sex) and the other sex would be heterozygous (the "heterogametic" sex). If there were recombination between the chromosome pair in the heterogametic sex, the gene might switch chromosomes, thus converting an X chromosome to a Y chromosome and the corresponding Y chromosome to an X chromosome. The status quo would be preserved and equal numbers of differentiated gametes would still be produced.

However, if the complexities of sexual differentiation were to require more than one gene, -- say genes X₁ and X₂ on the X chromosome, and the corresponding allelic genes Y₁ and Y₂ on the Y chromosome, -- the situation would get more complicated. Recombination (crossing-over between chromosomes at meiosis) might then separate the two genes to generate chromosomes (and hence gametes) with genes X₁ and Y₂ together, and genes Y₁ and X₂ together. Sexual differentiation would be impaired.

To prevent this happening either the gene pairs, -- X₁ and X₂, and Y₁ and Y₂ -- would have to be closely linked on the X and Y chromosomes, respectively, and/or one of the chromosomes would have to develop some local mechanism to prevent recombination (Ohno 1967). If such antirecombination activity could not easily be localized, then the activity might spread to involve other genes on the chromosomes; these genes might themselves play no role in sexual differentiation.

This seems to be the situation that often prevails. The predominant function, sex determination, overrules, but does not necessarily eliminate, the functions of other genes on the same chromosome. The chromosomes are referred to as the sex chromosomes, even though concerned with many functions not related to sex.

Recombination has evolved because it is advantageous (Winge, 1917; Dougherty, 1955; Click Here). The benefits of recombination are not to be lightly discarded. However, for the initiation of speciation, recombination is a hazard; indeed, the prevention of recombination between members of different incipient species, or of an incipient species and the parental species ("reproductive isolation"), appears to be a fundamental part of the speciation process (Forsdyke, 1996, 1998, 1999a, b).

On the other hand, sex chromosomes exist within a species, and antirecombination is only beneficial to the extent that it prevents recombination between genes specifically concerned with sexual differentiation. If antirecombination activity should spread to encompass genes not concerned with sexual differentiation, the latter would loose any benefits recombination might have conferred. To the extent that recombination allows the correction of damaged genes (Bernstein & Bernstein, 1991), any non-lethal mutations resulting from such damage would remain uncorrected. There would then be no prevention, within species, of further changes in sex chromosomes, including additions, deletions and other "macromutations" similar to those that occur between species as they differentiate (Chandley et al. 1975).

Thus, instead of remaining homomorphic, the sex chromosomes might come to differ in size (heteromorphic). In some species (e.g. certain fish and amphibia) the sex chromosomes remain homomorphic and appear to differ only in the genes affecting sexual differentiation. In other species (e.g. mammals, birds) the sex chromosomes are heteromorphic. It is argued that modern heteromorphic sex chromosomes evolved from homomorphic ancestors (Ohno 1967).

As noted above, species and sexual differentiations have an important feature in common. For their maintenance there must be an absence of recombination, either between one or more chromosome pairs in both sexes (speciation), or between regions of one chromosome pair in one sex (sexual differentiation).

For species differentiation, chromosomes of members of a variant group must not recombine with those of the parental line, otherwise character differences due to multiple genes could blend, and the differentiation would be lost.

For sexual differentiation, regions of the sex chromosomes must not recombine because the characters conferring sexual identity could blend, and sexual differentiation would then be lost. The sex chromosomes (e.g. X and Y in humans, where the male is the heterogametic sex) must, between themselves, maintain something akin to the reproductive isolation which defines organisms as members of distinct species (Forsdyke, 1999b).

The most obvious features of both differentiations are those due to differences in individual gene products, which might include gene products affecting recombination. This has led to a "genic" view of speciation (Coyne, 1992). An alternative "chromosomal" view postulates failure of recombination due to lack of homology between chromosomes attempting to pair at meiosis (Darlington, 1932; Gulick, 1932; White, 1978; King, 1993). Several objections to this view (Coyne & Orr, 1998) have been addressed in recent work (Forsdyke, 1996, 1998, 1999a, b). Further support for the chromosomal view would be provided if it could explain, more economically than the genic alternative, the phenomenon known as "Haldane's rule".

Indeed, if we were to go back through time (upwards, along the arms of our inverted Y genealogy), it is likely that we would arrive at proto-horses and proto-giraffes which were able to copulate. There would be no transfer barrier separating male gametes from female gametes, and fertilization might then occur so that the parental genomes could come to coexist in one cell (zygote). Thus, there would be no prezygotic reproductive isolation. However, by this time after the initial divergence the species differentiations would have been so advanced that the products encoded by the two genomes would not be able to cooperate to allow development to occur. Thus, the hybrid would not survive. The transfer barrier might not have yet arisen, but this developmental barrier (producing "hybrid inviability") would still have ensured reproductive isolation.

Nevertheless, the developmental barrier is unlikely to be responsible for the initial branching event in most cases of species formation. If we were to go even further back towards the time of the initial divergence, we would probably encounter proto-horses and proto-giraffes which were able both to copulate successfully (no transfer barrier), and to give rise to offspring which grew to become healthy, vigorous, adults (no developmental barrier). At this time, although the transfer and developmental barriers had not yet arisen, there would still have been reproductive isolation. The hybrids, although vigorous, would have been sterile ("mules"), and so no offspring would be produced.

Thus, moving further up the arms of our inverted Y genealogy, we encounter the barrier to reproduction most proximate to the point of species divergence ("hybrid sterility"). We are very close to the branch point leading downwards to proto-horses and proto-giraffes, but we are not quite at it. Hybrid sterility is not an all-or-none phenomenon. Initially hybrid sterility is only partial, so that some fertile offspring are produced. The first manifestation of a species branching into two lines is a degree of hybrid sterility. When examining the offspring of a cross between such lines a remarkable regularity became apparent to a young mathematically-gifted biologist whose mentors included Bateson (Haldane, 1957).

John Burdon Sanderson Haldane (1892-1964). Photograph prepared by Dr. Martin Case from a passport photograph. This is the frontpiece of Ronald Clark's 1968 biography published by Hodder and Stoughton, London.

Although it has attracted the attention of biologists for centuries, J. B. S. Haldane in 1922 appears to have been the first to enunciate that "when in the F1 [first generation] offspring of two different animal races [lines], one sex is absent, rare or sterile, that sex is the heterozygous [heterogametic] sex" (Craft, 1938). The absence of one sex among the offspring can reflect a failure of the fertilized ovum (zygote) to develop properly. This is Haldane's rule for "hybrid inviability" (Forsdyke, 1995). We are here concerned primarily with Haldane's rule for "hybrid sterility"(Fig. 2).

JBS Haldane

FIG. 2. Progression towards complete (sympatric) reproductive isolation between two "lines", such that they become "species".

In the normal situation (top) there are equal numbers of males and females among the progeny of a cross, and all are fertile (F).

Incipient speciation is marked by preferential loss of fertility (sterility; S) in the heterogametic sex (in this case males).

Reproductive isolation is complete when this progresses to hybrid sterility in all offspring (not necessarily accompanied by a decrease in number of offspring).

This hybrid sterility barrier may be replaced by hybrid inviability (bottom: initially fewer viable offspring, progressing to no viable offspring).

In the general case (initially no allopatric isolation) these two post-zygotic barriers might eventually be replaced by prezygotic barriers (not shown; e.g. inability to copulate, geographic isolation).

Haldane's rule usually concerns crosses between members of different "races" (varieties, breeds, lines) within what appears as a single species (defined reproductively; Forsdyke, 1999b). We are accustomed to recognize as "races," groups within a species the members of which show some common morphological differentiation from the members of other groups. In principle, members of a group ("race") within a species might begin to differentiate with respect to reproductive potential (no longer retaining full fertility when crossed with parental stock) before any morphological or other physiological differences are evident (Romanes, 1886; 1897). When fully differentiated in this way, the members of the group would be reproductively isolated, and so would constitute a distinct species, even if not anatomically or functionally distinguishable from members of the parental group.

Thus, Haldane's rule has no obligatory anatomical or physiological correlates, but is a phenomenon of incipient speciation alone; as such it is regarded as having the potential to tell us much about this process (Coyne, 1992; Naveira & Maside, 1998). It is proposed (Forsdyke, 1999b) that the preferential sterility of the heterogametic sex can be regarded as a step, or way station, on the path towards the complete sterility associated with full species differentiation, a process generally accompanied by the anatomical and physiological differentiation of conventional phenotypic characters.

In the homogametic sex the two sex chromosomes are essentially identical (e.g. genotype XX in human females), and can recombine at meiosis. The opportunity for different sex chromosomes to recombine at meiosis arises in the heterogametic sex (genotype XY in human males). Here, as in the homogametic sex, there would be a homology search perhaps initiated by "kissing" interactions between the aligned chromosomes (Forsdyke, 1996, 1998; Kleckner, 1997; Gupta, Folta-Stogniew & Radding, 1999; Zaitsev & Kowalczykowski, 1999). If this search were successful, double-strand breaks and recombination would occur. For the regions of the sex chromosomes concerned with sexual differentiation to become "reproductively isolated", it might be necessary for the homology search to fail in these regions, either through changes in specific genes affecting the process (a "genic" cause of non-recombination; Dobzhansky, 1933; Coyne & Orr, 1998), or through lack of homology (a "chromosomal" cause of non-recombination; Darlington, 1932; Gulch, 1932; White, 1978; King, 1993).

The sex chromosomes tend to become progressively differentiated from each other, and in some species recombination remains possible only in small "pseudoautosomal" regions (Koller & Darlington, 1934). The non-pseudoautosomal regions of sex chromosomes appear to be kept from pairing and recombination by a failure of homology (a "chromosomal" hypothesis), rather than by a failure of hypothetical gene products specifically dedicated to this pairing, but not to the pairing of the autosomes and pseudoautosomal regions (a "genic" hypothesis).

While, at the initiation of sex chromosome differentiation it is possible that there might have been early transient genic incompatibilities which would later have been superseded by secondary "chromosomal" incompatibilities, current opinion favours the latter as the primary cause of the failure of the sex chromosomes within a species to recombine (Graves, Disteche & Toder, 1998; Graves & O'Neill, 1997; Guttman & Charlesworth, 1998; Steinemann & Steinemann, 1998, 1999).

If the mechanisms of sexual and species differentiations were similar (and we here take the similarities to be both chromosomal), then, since the opportunity for recombination between different sex chromosomes (e.g. X and Y) can occur only in the sex with both chromosomes (the heterogametic sex), by preventing recombination that sex could be considered to have taken a step towards speciation. Whereas, for species differentiation, the homogametic sex would have to differentiate both sex chromosomes and autosomes, the heterogametic sex would have only the autosomes to differentiate. By virtue of this head-start, the heterogametic sex would be rare or absent among the progeny of crosses between an incipient species and its parental stock (Haldane's rule).

The latter sentence is in error. It should read: "By virtue of this head-start, the heterogametic sex would be sterile (Haldane's rule for hybrid sterility)."

Apologies DRF Aug 2003

On the other hand, it is possible that the chromosomal bases of species and sexual differentiations are mechanistically quite distinct. The process which prevents the chromosomes of members of a variant group within a species from recombining with those of the parental line (thus providing the "intrinsic" isolation fundamental to many forms of incipient speciation; Romanes, 1886, 1897; Forsdyke, 1996, 1998, 1999a,b), might be quite unrelated to the process which prevents an X chromosome recombining with part of a Y chromosome.

In this respect it should be noted that the differentiation of parts of the sex chromosomes within a species clearly does not impair gametogenesis in the heterogametic sex, whereas differentiation of sex chromosomes and autosomes in the course of speciation impairs gametogenesis in hybrids of both sexes (hybrid sterility). Thus, the "checkpoint" leading to meiotic disruption and impaired gametogenesis would have to be inactive in the case of sex chromosome differentiation at least until such a time as the sex chromosome differentiation came to contribute to species differentiation.

To complete the speciation process the meiotic homology search must fail both between the paired pseudoautosomal regions of the sex chromosomes, and between paired autosomes. The sterility accompanying this has been recognized for thousands of years in the form of the mule.

mule, the vigorous, but sterile, product of a cross between horse and ass

The cross between a horse and an ass is productive (first crosses succeed) producing a mule. However, the mule is sterile so that a subsequent cross would fail (second crosses fail). The sterility is associated with a degenerate gonad (Stephan, 1902; Guyer, 1902; Chandley et al., 1974), presumably produced by "check-point" mechanisms which detect errors in meiosis (Page & Orr-Weaver, 1996; Smith & Nicolas, 1998).

A case has been made that the failure of pairing of homologous chromosomes at meiosis is of "genic" origin (Coyne & Orr, 1998). Indeed, the pairing of homologous chromosomes would require some gene products, and thus mutations in these genes should impair pairing. Such genes might well be identified, mapped and successfully cloned. However, the validity of some experiments claiming precise localisation of genes having a major effect on hybrid sterility has been questioned (Maside & Naveira, 1996; Naveira & Maside, 1998).

The "chromosomal view" sees homologous chromosomes normally pairing or "conjugating" due to complementary features, likened to "the sword and the scabbard". Bateson (1922b) wondered how this complementarity might be lost:

In support of this essentially chromosomal viewpoint (although the authors refer to it as "polygenic"), it has been shown that small differences in base composition, a known species-specific characteristic, should suffice to disrupt pairing (Forsdyke, 1996, 1998, 1999a,b). Such "micromutations", which change base composition (the percentage C+G), would not be observed microscopically, but would function collectively to disrupt pairing in the fashion suggested by Naveira and Maside (1998).

Later-developing secondary mechanisms of isolation, substituting for the primary micromutation-dependent mechanism, would include prezygotic isolating factors (e.g. mating preferences constituting a transfer barrier), and chromosomal macromutations. A strength of the micromutation version of the chromosomal hypothesis is that it allows fine degrees of compatibility, so that a rare reproductive variant would be more likely to find a partially matching "physiological complement" (Romanes, 1886; 1897).

Further evidence for the chromosomal view derives from studies in allopolyploids. Here, sterility due to genic incompatibilities should usually still be manifest, whereas the presence of a pairing partner should effectively "cure" sterility due to chromosomal incompatibilities. This polyploidy test (Winge, 1917) can discriminate between genic and chromosomal causes of hybrid sterility (Dobzhansky, 1937).

Some applications of the polyploidy test by Dobzhansky (1933) show a genic cause to apply in his particular cases of sterility. Other applications show that sterility may also have a chromosomal basis (Darlington, 1932; Dobzhansky, 1937). These results are consistent with cytological studies of gametogenesis in hybrid mice (Hale et al., 1993; Matsuda et al., 1991, 1992).

From the chromosomal viewpoint it follows that for speciation the homogametic sex has to complete three steps,

1) differentiation of the sex chromosomes,

2) differentiation of autosomal pairs, and

3) activation of "check-points" (Page & Orr-Weaver, 1996) which respond to such differentiations by disrupting meiosis.

In contrast the heterogametic sex, being already advanced in the first step, has essentially to complete only the latter two steps. Thus, incipient speciation, manifest as hybrid sterility, should be most apparent in the heterogametic sex.

Many observations in the Haldane's rule literature can be reassessed from this perspective. For example, recent studies of Presgraves and Orr (1998) showed that mosquito genera with homomorphic sex chromosomes display hybrid sterility, but not hybrid inviability. This is consistent with the above hypothesis, and is also consistent with hybrid inviability being based on differences in dosage compensation (which probably requires heteromorphic sex chromosomes; Forsdyke, 1995). The proposed chemical basis for the primary chromosomal change (differences in (C+G)%; Forsdyke, 1996, 1998), lends itself well to the present model.

Although some members of the genic school believe that "the main causes of Haldane's rule now seem clear", it is admitted that "many ancillary questions await further analysis" (Turelli, 1998). Presented as an economical explanation for Haldane's rule, the arguments set out above for the chromosomal hypothesis also provoke further questions.

When the speciation process approaches completion, the pairing of all chromosomes is impaired at meiosis in hybrids of both sexes. Yet, if the requirements of sexual differentiation can be satisfied by one "way station", namely development of incompatibility in only one chromosome pair (the sex chromosomes), why should the initial requirements of species differentiation not be satisfied by another "way station", namely development of incompatibility in only one autosome pair? This should suffice to activate check-point mechanisms and disrupt gametogenesis. Must the initiation of speciation necessarily be accompanied by differentiation of all the autosomes?

The results of Naveira and Maside (1998) suggest not. A second "way station" involving only one autosome pair should indeed suffice to complete the speciation process. Under the shelter of the reproductive isolation so afforded, the remaining chromosomes should rapidly develop disparities (both micromutations and macromutations). It seems unlikely that the meiotic checkpoint(s) would require activation by signals from several chromosomes for gametogenesis to be disrupted.

It should be noted that the homology search mechanism, perhaps needing coordinated stem-loop extrusions under negative supercoiling (a long-range effect of locally-acting topoisomerases or transcription complexes; Forsdyke, 1996, 1998; Strick et al., 1998; Wong et al., 1998), implicates large chromosomal regions. In this way the initially-developing anti-recombination activity in homomorphic sex chromosomes might have spread and so facililitated the transition to heteromorphism.

I thank Harcourt-Brace Inc. for granting copyright permission to place full-text versions of relevant preceding papers at my web-site.

BATESON, W. (1904). Draft application to the Balfour Fund. In: William Bateson, F. R. S. Naturalist. His Essays and Addresses (Bateson, B., ed.), p. 94. Cambridge: Cambridge University Press, 1928.

BATESON, W. (1908). The methods and scope of genetics. Inaugural lecture at the University of Cambridge. In: William Bateson, F. R. S. Naturalist. His Essays and Addresses (Bateson, B., ed.), pp. 317-333. Cambridge: Cambridge University Press, 1928.

BATESON, W. (1909). Mendel's Principles of Heredity, pp. 164-195. Cambridge: Cambridge University Press.

BATESON, W. (1922a). Evolutionary faith and modern doubts. Nature 109, 553-556.

BATESON, W. & SAUNDERS, E. R. (1902) Report I. In: Reports to the Evolution Committee of the Royal Society, pp. 1-160. London: Harrison.

BERNSTEIN, C. & BERNSTEIN, H. (1991). Aging, Sex and DNA Repair. San Diego: Academic Press.

BUSH, G. L., CASE, S. M., WILSON, A. C. & PATTON, J. L. (1977). Rapid speciation and chromosomal evolution in mammals. Proc. Natl. Acad. Sci. USA, 74, 3942-3946.

CHANDLEY, A. C., JONES, R. C., DOTT, H. M., ALLEN, W. R. & SHORT, R. V. (1974). Meiosis in interspecific equine hybrids. 1. The male mule (Equus asinus X E. caballus) and hinny (E. caballus X E. asinus). Cytogen. Cell Genet. 13, 330-341.

COYNE, J. A. & ORR, H. A., (1998). The evolutionary genetics of speciation. Phil. Trans. R. Soc. B (London) 353, 287-305.

CRAFT, W. A. (1938). The sex ratio in mules and other hybrid mammals. Q. Rev. Biol. 13,19-39.

DARLINGTON, C. D. (1932). Recent Advances in Cytology, pp. 448-485. London: Churchill.

DARWIN, C. (1851). A Monograph on the Subclass Cirripedia, vol. 1, pp. 281-293. London: The Ray Society.

DARWIN, C. (1859). The Origin of Species by Means of Natural Selection. London: Murray.

DE VRIES, H. (1910). Intracellular Pangenesis (Gager, C. S., translator), pp. 18-19. Chicago: Open Court.

DOBZHANSKY, T. (1933). On the sterility of the interracial hybrids in Drosophila pseudoobscura. Proc. Natl. Acad. Sci. USA 19, 397-403.

DOBZHANSKY, T. (1937). Genetics and the Origin of Species, pp. 288-291. New York: Columbia University Press.

DOUGHERTY, E. C. (1955). Comparative evolution and the origin of sexuality. Syst. Zool. 4, 145-190.

FORSDYKE, D. R. (1995). Fine tuning of intracellular protein concentrations, a collective protein function involved in aneuploid lethality, sex-determination and speciation? J. theor. Biol. 172, 335-345.

FORSDYKE, D. R. (1996). Different biological species "broadcast" their DNAs at different (G+C)% "wavelengths". J. theor. Biol. 178, 405-417.

FORSDYKE, D. R. (1998). An alternative way of thinking about stem-loops in DNA: a case study of the G0S2 gene. J. theor. Biol. 192, 489-504.

FORSDYKE, D. R. (1999a). "The Origin of Species", revisited. Queen's Quarterly 106, 112-133.

FORSDYKE, D. R. (1999b). Two levels of information in DNA: relationship of Romanes' "instrinsic" variability of the reproductive system, and Bateson's "residue" to the species-dependent component of the base composition, (C+G)%. J. theor. Biol. 201, 47-61.

GRAVES, J. A. M., DISTECHE, C. M., & TODER, R. (1998). Gene dosage in the evolution and function of mammalian sex chromosomes. Cytogenet. Cell Genet. 80, 94-103.

GRAVES, J. A. M. & O'NEILL, R. J. W. (1997). Sex chromosome evolution and Haldane's rule. J. Hered. 88, 356-360.

GULICK, A. (1932). Evolutionist and Missionary, John Thomas Gulick, pp. 498. Chicago: University of Chicago Press.

GUPTA, R. C., FOLTA-STOGNIEW, E. & RADDING, C. M. (1999). Human Rad51 protein can form homologous joints in the absence of net stand exchange. J. Biol. Chem. 274, 1248-1256.

GUTTMAN, D. S. & CHARLESWORTH, D. (1998). A X-linked gene with a degenerate Y-linked homologue in a dioecious plant. Nature 393, 263-266.

GUYER, M. F. (1902). Hybridism and the germ cell. Bull. Univ. Cincinnati 21, 1-20.

HALDANE, J. B. S. (1922). Sex ratio and unisexual sterility in hybrid animals, J. Genet. 12, 101-109.

HALDANE, J. B. S. (1957). The theory of evolution, before and after Bateson. J. Genet. 56, 11-27.

HALE, D. W., WASHBURN, L. L. & EICHER, E. M. (1993). Meiotic abnormalities in hybrid mice of the C57BL/6J x Mus spretus cross suggest a cytogenetic basis for Haldane's rule of hybrid sterility. Cytogen. Cell Genet. 63, 221-234.

KING, M. (1993). Species Evolution. The Role of Chromosome Change. Cambridge: Cambridge University Press.

KLECKNER, N. (1997). Interactions between and along chromosomes during meiosis. Harvey Lect. 91, 21-45.

KOLLER, P. C. & DARLINGTON, C. D. (1934). The genetical and mechanical properties of the sex chromosomes. 1. Rattus norvegicus. J. Genet. 29, 159-173.

LAURIE, C. C. (1997). The weaker sex is heterogametic. 75 years of Haldane's rule. Genetics 147, 937-951.

MASIDE, X. R. & NAVEIRA, H. F. (1996). A polygenic basis of hybrid sterility may give rise to spurious localizations of sterility factors. Heredity 77, 488-492.

MATSUDA, Y., HIROBE, T. & CHAPMAN, V. M. (1991). Genetic basis of X-Y chromosome dissociation and male sterility in interspecific hybrids. Proc. Natl. Acad. Sci. USA 88, 4890-4854.

MATSUDA, Y., MOENS, P. B. & CHAPMAN, V. M. (1992). Deficiency of X and chromosome pairing at meiotic prophase in spermatocytes of sterile interspecific hybrids between laboratory mice (Mus domesticus) and Mus spretus. Chromosoma 101, 483-492.

NAVEIRA, H. F. & MASIDE, X. R. (1998). The genetics of hybrid male sterility in Drosophila. In: Endless Forms: Species and Speciation. (Howard, D. J. & Berlocher, S. H., eds.), pp. 330-338. Oxford: Oxford University Press.

OHNO, S. (1967). Sex Chromosomes and Sex-Linked Genes, pp. 1-23. New York: Springer-Verlag.

PAGE, A. W. & ORR-WEAVER, T. L. (1996). Stopping and starting the meiotic cycle. Curr. Opin.Genet. Devel. 7, 23-31.

PRESGRAVES, D. C. & ORR, H. A., Haldane's rule in taxa lacking a hemizygous X. Science 282, 952-954.

PUNNETT, R. C. & BATESON, W. (1908). The heredity of sex. Science 27, 785-787.

ROMANES, G. J. (1886). Physiological selection. An additional suggestion on the origin of species. J. Linn. Soc. Zool. 19, 337-411.

ROMANES, G. J. (1897). Darwin, and After Darwin. Vol. 3. Post-Darwinian Questions. Isolation and Physiological Selection. London: Longmans, Green.

SMITH, K. N. & NICOLAS, A. (1998). Recombination at work for meiosis. Curr. Opin. Genet. Devel. 8, 200-211.

STEINEMANN, M. & STEINEMANN, S. (1998). Enigma of Y chromosome degeneration: Neo-Y and neo-X chromosomes of Drosophila miranda a model for sex chromosome evolution. Genetica 102, 409-420.

STEINEMANN, S. & STEINEMANN, M. (1999). The Amylase gene cluster on the evolving sex chromosomes of Drosophila miranda. Genetics 151, 151-161.

STEPHAN, P. (1902). Sur la structure histologique du testicule des mulets. Compt. Rendu. Ass. Anat. 4, 37-46.

STRICK, T. R., CROQUETTE, V. & BENSIMON, D. (1998). Homologous pairing in stretched supercoiled DNA. Proc. Natl. Acad. Sci. USA. 95, 10579-10583.

TURELLI, M. (1998). The causes of Haldane's rule, Science 282, 889-891.

WHITE, M. J. D. (1978). Modes of Speciation, pp. 323-349. San Francisco: Freeman.

WINGE, O. (1917). The chromosomes, their number and general importance. Comp. Rend. Trav. Carls. Lab. 13,131-275.

WONG, B. C., CHIU, S-K. & CHOW, S. A. (1998). The role of negative superhelicity and length of homology in the formation of paranemic joints promoted by RecA protein. J. Biol. Chem. 273, 12020-12127.

ZAITSEV, E. N. & KOWALCZYKOWSKI, S. C. (1999). The simultaneous binding of two double-stranded DNA molecules by Escherichia coli RecA protein. J. Mol. Biol. 287, 21-31.

End Note (2005) Independent evidence for the above argument appeared in a paper by Ironside & Filatov (Genetics 171, 705-713). Sex chromosomes of certain species of the plant genus Silene are 'homomorphic' to the extent that X and Y share similar genes (i.e. genes are not hemizygous in the heterogametic sex), although there may be minor size differences between these chromosomes (see below). The authors found high divergence in non-coding regions of X and Y homologues, with differentiation between species being less for the X than for the Y homologues. It was concluded:

"The higher interspecific divergence of DD44Y, compared to DD44X, supports the hypothesis that Y chromosome differentiation between incipient species precedes reproductive isolation of the entire genome, forming an early stage in the process of speciation."

End Note (Dec 2010) Plants in the genus Silene with sex chromosomes follow Haldane's rule, so the rule applies both to animals and plants (Brothers & Delph, Evolution 64, 3643-48).

End Note (2016) Further evidence for the above argument appeared in a paper by Delph & Demuth (Journal of Heredity 107, 383-391) under the heading "Faster Heterogametic Sex Theory." It is concluded:

"There are 3 ways that simply being heterogametic may foster the evolution of Haldane's Rule. The first, originally favored by Haldane (1922), is that non-coevolved X and Y (or Z and W) may have difficulty with proper segregation during meiosis. Haldane later suggested that this explanation was "inadequate" and evidence from Drosophila mutants with Y to X chromosome translocations caused him to favor a second, more generic, suggestion that the rule is a consequence of "interspecific differences in the sex chromosomes" Haldane (1932). Since translocations from the Y to the X (or W to Z) in 1 species will result in heterogametic hybrids that are missing part of their genome, the heterogametic sex potentially evolves postzygotic isolation faster than the homogametic sex. Although the original chromosomal mechanism per se (i.e. mispairing of XY or ZW in hybrids) largely fell out of favor, recent work has argued for its importance on theoretical grounds (Forsdyke 2000) and it is the best explanation for male rarity in some Silene hybrids noted above (Demuth et al. 2014)."

Hu and Filatov (2016) found that a "large-X effect in plants" had "increased species divergence and reduced gene flow on the Silene X-chromosome," but gene flow involving autosomal loci was "sufficient to homogenize the gene pools of the two species." (Molecular Ecology 25, 2609-19). Armstrong and Filotov (2008) had noted the Y chromosome to be 38% larger than the X chromosome and considered:

"Thus, currently there is little evidence for active genetic degeneration of S. latifolia Y chromosome. Accumulation of repetitive 'junk' DNA on the S. latifolia Y chromosome reported by a number of studies (Hobza et al., 2006; Kejnovsky et al., 2006) cannot be taken as evidence for degeneration, as accumulation of 'junk' DNA on the Y does not automatically lead to gene loss or dysfunctionalisation.

The sex chromosomes were implicated in the initiation of divergence into species. The possibility that, irrespective of genic effects, accumulation of 'junk' DNA could impair mieotic pairing between X and Y chromosomes was not excluded. Whether this accumulation had been favoured, even earlier, by non-genic or genic differences in GC% (that might have impaired pairing) was not considered.

End Note (Feb. 2017) The "only a question of foreign DNA amount" argument of Naviera and Masade (1998) is supported by the independent fruitfly studies of Amanda Moehring (2011) who writes of "different amounts of heterospecific genome" and the mouse studies of Forejt and coworkers, who write (2013):

"The asynapsis may represent a universal mechanistic basis of F1 hybrid sterility manifested by pachytene arrest. It is tempting to speculate that a fast-evolving subset of the noncoding genomic sequence important for chromosome pairing and synapsis may be the culprit."

Moehring AJ (2011) Heterozygosity and its unexpected correlations with hybrid sterility. Evolution 65: 2621-2630.

Bhattacharyya et al. (2013) Mechanistic basis of infertility of mouse intersubspecific hybrids. Proc Nat Acad Sci USA 110:E468-E477

	"When in the F₁ offspring of two different animal races one sex is absent, rare, or sterile, that sex is the heterozygous [heterogametic] sex."
J.B.S. Haldane (1922) Sex-ratio and unisexual sterility in hybrid animals. J. Genetics 12, 101-109.

Thus in humans there are two types of spermatozoa, X-spermatozoa and Y-spermatozoa. If these are produced in equal quantitities by the heterogametic sex (the normal situation) then, on average, half the children will be male (XY), and half the children will be female (XX).
	The two types of human spermatozoa at surface of ovum

The hybrid *inviability* occurring by the mechanism proposed in 1995 (dosage compensation; Click Here), and by some other mechanisms, would not be expected in "species" of genera with homomorphic sex chromosomes (i.e. the sex-determining chromosomes appear similar).
Mosquitos of the genus Aedes fit this criterion. Presgrave and Orr (1998; *Science* 282, 952-954) found for crosses between Aedes "incipient species" that Haldane's rule was associated with male hybrid *sterility*, not inviability.
Plants with separate sexes (dioecious plants) and with homomorphic sex chromosomes also follow Haldane's rule regarding male sterility, not inviability (see below).



FIG. 1. Hybrid sterility as the form of reproductive isolation responsible for the initiation of speciation in most cases of species formation. For further details please see text.