William Bateson 1861-1926 Portrait in the collection of the John Innes Centre, Norwich.

Richard Goldschmidt. From the National Library of Medicine collection, Bethesda, MD.

D. R. FORSDYKE (2003) 
J. Biological Systems
11, 341-350
With copyright permission from World Scientific Publishing Co.

Sometimes a cross between two individuals that appear to belong to the same species produces a sterile offspring (i.e. their hybrid is sterile). Thus, the two individuals appear reproductively isolated from each other. If each could find a compatible mate, then new species might emerge.

    At issue is whether the form of hybrid sterility that precedes sympatric differentiation into species is, in the general case, of genic or non-genic origin. Several recent papers lend the authority of William Bateson to the genic hypothesis, referring to the "Bateson-Dobzhansky-Muller hypothesis." All these papers cite a 1996 paper that, in turn, cites a 1909 paper of Bateson. However, from 1902 until 1926 the latter espoused a non-genic hypothesis that today would be classified as "chromosomal." Analysis of Bateson's 1909 text (Click Here) reveals no recantation.

    Bateson's non-genic view was similar to that advanced by Richard Goldschmidt in the 1940s. However, Bateson proposed a contribution from parents of abstract factors that, together in their hybrids, complement to bring about a negative effect (hybrid sterility).  In contrast, Goldschmidt proposed that normally parents contribute complementary factors making parental chromosomes compatible at meiosis in their hybrids, which hence are fertile (i.e. the parental factors work together to produce a positive effect). When the factors are not sufficiently complementary the parental chromosomes are incompatible in their hybrids, which hence are sterile. The non-genic Batesonian-Goldschmidtian abstractions are now being fleshed-out chemically in terms of DNA base-composition differences.

Keywords: Complementary factors; genic hypothesis; hybrid sterility; non-genic hypothesis; speciation

1. Introduction

2. The Species Problem

3. Complementary Factors are Not Genic

4. Genic Hypothesis of Dobzhansky and Muller

5. Goldschmidt's Non-Genic Hypothesis

6. Complementary Factors Make Chromosomes Compatible 

7. What are the Complementary Factors?

1. Introduction

I here continue a theoretical study of  "non-genic" speciation that built on the work of Romanes, Bateson and Haldane [15, 16]. I clarify a misinterpretation of Bateson, extend the study to the work of Goldschmidt, and show how the latter relates to earlier work of Bateson and Romanes.

2. The Species Problem

William Bateson brought the work of Gregor Mendel to the attention of the English-speaking world, and gave us words such as "allelomorph" (i.e. allele), "epistasis," "homozygote," "heterozygote," and "homeotic." In the early decades of the 20th century Bateson held a chair at Cambridge University, and became the first head of the John Innes Horticultural Institute [18]. While he and his coworkers made many contributions to genetics, the focus of his work was "the species problem" - namely, how did one lineage diverge into two, reproductively-isolated, lineages? On this question he developed a consistent view that, although repeatedly expressed in numerous books and papers, failed to convince his contemporaries and many who came after. In the words of one modern commentator he may have been "a biologist ahead of his time" [1].

    A recent paper by Navarro and Barton [29] describes "a new class of models" of speciation that suggest "chromosomal changes are strong genetic barriers because they reduce recombination between heterokaryotypes, - - such strong barriers would facilitate divergence."  This is entirely supportive of Bateson's viewpoint. However, it is noted that the effects of the barriers "would be especially pronounced if divergence is through the accumulation of incompatible alleles, as proposed by Bateson, Dobzhansky, and Muller." This will have suggested to many readers that Bateson's view was the same as that of the eminent geneticists Herman J. Muller and Theodosius Dobzhansky. To support this there is a single reference to a paper by geneticist H. Allen Orr [31]. That paper, in turn, cites a paper presented by Bateson at the 1909 centenary celebration of the birth of Charles Darwin [2]. What view did Bateson express in the latter paper? Did Orr accurately relay it?


3. Complementary Factors are Not Genic

Partial hybrid sterility is an early manifestation of a sympatric divergence within a species that can lead to full sterility and hence to the reproductive isolation which defines species (for reviews see refs. [17, 33]). Typically, the defect is in the gonad where, presumably following activation of a meiotic "check-point" [32], gamete production fails [22]. Arguing from first principles, Bateson held that, since neither of the parents of a partially or fully sterile hybrid were themselves sterile, they could not individually have possessed, in an active form, a factor causing sterility [2]. Rather, each parent would have donated one of two "complementary factors" that would collectively need to be active within the hybrid for the production of sterility. To this extent the parents could be considered as complementary individuals distinct from other potentially pairing members of the species. 

    Thus, in a species consisting of males A-D, and females E-H, A and E might produce sterile offspring, but fertile offspring would result from all other possible crosses (e.g. between A and F, G, or H, between B and E, F, G or H, etc.). Only complementary individuals would donate complementary factors, which would then "arrest or prevent [meiotic] cell division" within the hybrid, but "need not, and probably would not, produce any other perceptible effects." These points were reiterated in Problems of Genetics, Bateson's last major work [3].

    Orr accurately quotes the 1909 paper: "When two species, both perfectly fertile severally, produce on crossing a sterile progeny, there is a presumption that the sterility is due to the development in the hybrid of some substance which can be formed only by the meeting of two complementary factors."  Thus the two complementary factors not only must be present in the hybrid, but also must meet in the hybrid in order to develop "some substance" through which they produce a negative effect, hybrid sterility. Although not quoted by Orr, to support this Bateson continued:

"That some such account is correct in essence may be inferred from the well-known observation that if the hybrid is not totally sterile but only partially so, and thus is able to form some good germ cells which develop into new individuals, the sterility of these daughter-individuals is sensibly reduced or may be entirely absent. The fertility once re-established, the sterility does not return in the later progeny, a fact strongly suggestive of segregation."

By this Bateson implied that a partially fertile hybrid (AE) from a cross between A and E is likely, on average, to find a mating partner within the population (A-H) with which it would be able to produce children (i.e. it would be fertile). Bateson's continuing discussion is quoted in-full by Orr as "remarkably prescient:"

"Now if the sterility of the cross-bred be really the consequence of the meeting of two complementary factors, we see that the phenomenon could only be produced among the divergent offspring of one species by the acquisition of at least two new factors; for if the acquisition of a single factor caused sterility the line would then end. Moreover each factor must be separately acquired by distinct individuals, for if both were present together, the possessors would by hypothesis be sterile. And in order to imitate the case of species each of these factors must be acquired by distinct breeds. The factors need not, and probably would not, produce any other perceptible effects -- . Not till the cross was actually made between the two complementary individuals would either factor come into play, and the effects even then might be unobserved until an attempt was made to breed from the cross-bred."

Orr next quotes a proposed experimental test. Bateson suggested obtaining:

"a pair of parents [one from each breed] which are known to have had any sterile offspring, and to find the proportions in which these steriles are produced. If, as I anticipate, these proportions are found to be definite, the rest is simple."

From this Orr draws the conclusion that Bateson espoused a genic hypothesis because he "expects such crosses will show that sterility is due to Mendelizing factors and, further, to pairs of interacting factors."

    It is here, I believe, that Orr goes seriously wrong. Because Bateson suggested his "factors" might show some of the properties of genes in that they might segregate, it does not follow that he considered his factors are either genes, or gene products. Bateson's caveat that his is a model for hybrid sterility, rather than for hybrid inviability, is dismissed by Orr as a "stumble."  In his writings Bateson carefully distinguished what we would now call "genic" from "non-genic" causes of hybrid sterility. Due to the random nature of mutation, genic causes would be unpredictable ("sporadic"), and would not reflect an underlying "remoteness of kinship" between pairing individuals. On the other hand, the partial or complete hybrid sterility resulting from the crossing of two particular individuals (A and E) was highly predictable. All AE offspring, whenever produced, are partially or completely sterile. Thus, with respect to the withered anther phenotype that was associated with poor or absent pollen production in plants, Bateson and his coworkers noted [7]:

"[Withered] anthers were seen from time to time in many families, though commonly confined to individual flowers. This sporadic sterility has not been particularly studied. It is of interest to compare this example of the definite appearance of sterility -- with the familiar [regular] occurrence of sterility in cross-breds [hybrids]. Such a phenomenon has often been supposed to indicate remoteness of kinship, yet here [in the case of sporadic sterility] a closely comparable effect occurs -- as the result of a cross between two types which must be nearly related."

   Today we recognize that many gene products are required for meiosis (e.g. for the activation of meiotic check-points). A chance mutation in any of these genes could result in sporadic sterility that could affect any hybrid, whether derived from parents of close kinship (producing hybrids AF, AG, AH, BE, BF, BG, BH, etc.), or less close kinship (producing hybrid AE).


4. Genic Hypothesis of Dobzhansky and Muller

Decades later, Dobzhansky [13] and Muller [28] suggested that hybrid sterility, of a type that could be widely involved in the reproductive isolation associated with speciation, was due to epistatic incompatibilities between gene products. They did not cite Bateson in this respect, presumably since they knew that, from the time of his major paper with Edith Saunders in 1902 [6] until his death in 1926, Bateson had ascribed species formation to a non-genic inherited "residue" that, unlike genes, was not subject to classical natural selection.

Furthermore, Bateson had always maintaining a high level of abstraction by the deliberate use of the non-committal term "factor." Thus, J. B. S. Haldane, who was much influenced by Bateson, later wrote [23] that he "never accepted the word 'gene' with its rather wide connotations -- Bateson used the neutral word 'factor' -- [which] can be anything from a difference of a few atoms in a single nucleotide, to an inversion or the presence of an extra chromosome; since these too are inherited in a Mendelian manner." In 1924  Bateson summarized [5]:

"We have determined the transferable [genic] characters as one group, and we no longer confound them with the essential elements conferring specificity [speciation]. Segregation [of genic characters] is of course often seen in species crosses, but as to the behaviour of these critical elements [conferring species specificity] we know as yet very little."

    In his 1996 paper Orr argues that Bateson offered the "Dobzhansky-Muller model" as early as 1909 and that there are "good reasons for thinking" neither of these authors knew of Bateson's prior model and that, in any case, they would have believed him as "irredeemably confused about evolution," and as "one of the enemies battled against" [31].  With respect to speciation Orr concludes that "Bateson and Dobzhansky not only arrived at the same conclusion, but at the right conclusion." I have argued that Dobzhansky's conclusion was neither right, nor the same as Bateson's [15-17]. Albeit only temporarily, it is conceivable that Bateson had recanted in his 1909 paper, but close examination of his text reveals no such wavering.

Nevertheless, the authority of Bateson has become increasingly linked to the genic hypothesis of Dobzhansky and Muller. Gathering momentum, Orr's argument is endorsed by his mentor [10] and relayed by Stewart Berlocher [8] in the opening chapter of the influential multi-author text Endless Forms: Species and Speciation. The argument is even dignified with an acronym "BDM," standing for the "Bateson, Dobzhansky, Muller model of speciation" [27]. Citing Orr, Norman Johnson [24] asserts that the epistatic genic model was "established by Dobzhansky and Muller and anticipated by Bateson." Thus, the new paper of Navarro and Barton [29], while itself tending to favour a "new class" of non-genic model, adds fuel to a fire that may already be out of control.


5. Goldschmidt's Non-Genic Hypothesis

While Dobzhansky [13] and Muller [28] were developing their genic speciation hypothesis, Richard Goldschmidt was developing a non-genic speciation hypothesis, which can now be seen as belonging to the class of "chromosomal" hypotheses [9, 21, 25, 36]. Sadly, "Goldschmidt has been ensconced in the annals of biological history as the originator of the idea of the hopeful monster, an idea that in reality represented only a minor component of his treatment of macroevolution" [34]. Much of an issue of the journal Paleobiology was devoted to largely negative reviews of a reprinting (probably inspired by S. J. Gould who wrote the preface), of Goldschmidt's 1940 classic, The Material Basis of Evolution (e.g. ref. [35]).

Goldschmidt distinguished Mendelian-type mutations, which he called "micromutations," from mutations concerned with divergence into biological species, which he called "systemic mutations" and considered would affect entire chromosomes. Thus he distinguished linear, "genic," species-maintaining, evolution ("microevolution") from branching, "non-genic," species-forming, evolution ("macroevolution"). Systemic mutations could be sub-microscopic: "the visible differences in regard to chromosomes are just one morphological character by which different species may or may not be recognized. It is also clear that these visible differences are not necessary features of evolution, as they may be completely absent." (see ref. [20], p. 186).

Macroevolutionary differences were held to be changes in chromosome "pattern" which would not necessarily affect genes. Goldschmidt wrote (see ref. [20], p. 191): "Within the species, the internal chromosome pattern may slowly change in a series of steps without any visible effect on the phenotype and without any accumulation of so-called gene mutations, small or large!" Fundamental to Goldschmidt's thinking, derived from consideration of sex chromosomes and polyploids [17], was the idea that each chromosome constituted a "reaction system" which could undergo "repatterning." Thus: "A repatterning of a chromosome may have exactly the same effect as an accumulation of [genic] mutations. And even more, a complete repatterning might produce a new chemical system --". Indicating his disagreement with Dobzhansky and Muller, he added: "This encourages me to believe that the dead end reached by neo-Darwinian theory based upon the conceptions of classical genetics can now be passed successfully." (see ref. [20], p. 203).

Along the lines of Bateson in 1914, who spoke of hereditary information as a "phenomenon of arrangement" [4], Goldschmidt attempted to give his concept of chromosomal "pattern" a chemical meaning (see ref. [20], p. 248):

"Let us compare the chromosome with its serial order to a long printed sentence made up of hundreds of letters of which only twenty-five different ones exist. In reading the sentence a misprint of one letter here and there will not change the sense of the sentence; even the misprint of a whole word (rose for sore) will hardly impress the reader. But the compositor might arrange the same set of type into a completely different sentence with a completely new meaning, and this in a great many different ways, depending upon the number of permutating letters and the complexity of the language --. To elevate such a model to the level of a biological theory we have, or course, to restate it in chemical terms."

This was published four years before Oswald Avery and coworkers produced evidence that DNA was the form in which hereditary information was transferred through the generations, ten years before Erwin Chargaff observed that (G+C)% was species-variant, and thirteen years before the determination of DNA structure [19, 30]. Goldschmidt proposed that (see ref. [20], p. 245) that:

"If only the serial pattern [of the chromosome] as a whole is decisive, an unlimited number of patterns is available without a single qualitative chemical change in the chromosomal material [modern interpretation: no change in DNA base composition], not to speak of a further unlimited number after qualitative changes (model: addition of a new amino acid into the pattern of a protein molecule) [modern interpretation: DNA base addition or substitution]. -- ."


6. Complementary Factors Make Chromosomes Compatible

At this point Goldschmidt departed from Bateson's proposal [2] of a contribution from parents of factors that, together in the hybrid, would complement to bring about a negative effect, namely hybrid sterility.  Goldschmidt proposed the converse, namely that normally parents contribute complementary factors making parental chromosomes compatible at meiosis in their hybrid, which hence would be fertile (i.e. the parental factors work together to produce a positive effect); when the factors were not sufficiently complementary the parental chromosomes would be incompatible in their hybrid, which hence would be sterile (i.e. the parental factors do not work together to produce a positive effect, so the default is a negative effect). Thus Goldschmidt wrote (see ref. [20], p. 245): "These pattern changes may be an accident, without any significance except for creating new conditions of genetic isolation by chromosomal incompatibility-- ."                   

Remarkably, Darwin's research associate, George Romanes, made a similar suggestion in 1886 postulating some intrinsic "peculiarity" in the gonad such that only individuals whose parents had been "physiological complements" would be fertile [15]. Apparently unaware of Romanes' work, Bateson had noted in 1913 that organisms could be incompatible either because the reproductive system of  "each is lacking in one of two complementary elements, or that each possesses a factor with an inhibitory effect" [3].

Romanes also clarified a point upon which his successors tended to be vague. Batesonian factors had to be "acquired by distinct breeds" [2]. This implied a necessity for some preexisting phenotypic differention into breeds, perhaps as the result of some degree of geographic isolation (e.g. the races of man, or varieties of a plant). Romanes spelled out that, in the general case, reproductive isolation of "intrinsic" origin must precede phenotypic differention, not the converse. The branching into two reproductively isolated lines would create conditions favouring subsequent anatomical and physiological differentiations. These differentiations might lead to hybrid inviability and mating incompatibilities. Such conditions could reinforce and eventually replace the originating cause of branching. This is recognized by Navarro and Barton [29] who note that the genetic barriers to recombination resulting from chromosomal changes "would facilitate divergence" which "would be especially pronounced" (my italics) if the phenotypic divergence were "through the accumulation of incompatible alleles." Thus, they envision the operation of genic, linear, Muller-Dobzhansky events that further distinguish species after the operation of non-genic, branching, Bateson-Goldschmidt events that originate species.

7. What are the Complementary Factors?

My own work, which is opposed by the genic school [26], has suggested that the intrinsic "peculiarity" of Romanes, the "complementary factors" of Bateson and the "patterns" of Goldschmidt are none other than Erwin Chargaff's species-dependent component of the base composition. Differences in the (G+C)% pattern would impede the pairing of chromosomes at meiosis, resulting in hybrid sterility. This internal non-genic barrier may be the most usual form of reproductive isolation permitting phenotypic differentiation within a species without geographic isolation, and with or without the influence of natural selection. Species arrival may or may not be immediately followed by genic differentiation for species survival, although natural selection will play a role in later phenotypic differentiation [17].

Thus, Bateson's abstract "factors" may be the sequences of homologous chromosomes that, when they "meet" in the hybrid, can be regarded as speaking with a particular (G+C)% "accent" [17], or having a particular (G+C)% "colour" (S-J. Lee, J. R. Mortimer and D. R. Forsdyke, unpublished work; Click Here ). When accent or colour match, then pairing occurs (i.e. "some substance" is developed in the hybrid) and meiosis progresses normally. When accent or colour are incompatible, then pairing does not occur (i.e. "some substance" is not developed in the hybrid), and meiosis fails so that hybrids are sterile. Thus, Bateson's abstraction ("some substance") can be equated with the meiotic pairing of homologous sequences from corresponding maternally-derived and paterally-derived chromosomes.

Referring to the battle between the "biometricians" and Bateson at the turn of the nineteenth century, Orr concedes that "we evolutionists" displayed an "infamous reluctance to surrender Galton in the face of Mendelism" [31]. A century later we are perhaps seeing a similar reluctance on the part of mathematical population geneticists to surrender Muller-Dobzhansky in the face of what might once have been "some vague principle which assiduously escapes all attempts to define it more clearly" [14], but is now being fleshed out in chemical terms.

Darwin, the initiator of the modern era of speciation research [11], wrote in 1874 that: "False facts are highly injurious to the progress of science, for they often endure long; but false views, if supported by some evidence, do little harm, for everyone takes a salutary pleasure in proving their falseness" [12]. Unfortunately, this view may itself be false. Sadly, everyone may not take pleasure in proving false a view that has abstract elements, and/or is politically correct. As has been shown here, what begins as a "view" can easily mutate into a "fact" which, as Darwin noted, has the potential to "endure long."



Queen's University hosts my web-pages where full-text versions of several references may be found.



[1]   Bateson P., William Bateson: a biologist ahead of his time, J. Genetics 81 (2002) pp. 49-58.

[2]   Bateson W., Heredity and variation in modern lights. In Darwin and Modern Science, ed. by Seward A. C. (Cambridge University Press, Cambridge, 1909) pp. 85-101. (Click Here)

[3]   Bateson W., Problems of Genetics (Yale University Press, New Haven, 1913) pp. 238-241.

[4]   Bateson W., Presidential address to the British Association, Nature 93 (1914) pp. 635-642.

[5]   Bateson W., Progress in biology, Nature 113 (1924) pp. 644-646.

[6]   Bateson W. and Saunders E. R., Report 1. Reports to the Evolution Committee of the Royal Society (Harrison, London, 1902) pp. 148-149. (Click Here)

[7]   Bateson W., Saunders E. R. and Punnett R. C., Report 2. Reports to the Evolution Committee of the Royal Society (Harrison, London, 1904) pp. 1-131.

[8]   Berlocher S. H., A brief history of research on speciation. In Endless Forms: Species and Speciation, ed. by Howard D. J. and Berlocher S. H. (Oxford University Press, Oxford, 1998) pp. 3-15.

[9]   Bush G. L., Case S. M., Wilson A. C. and Patton J. L., Rapid speciation and chromosomal evolution in mammals, Proc. Natl. Acad. Sci., USA 74 (1977) pp. 3942-3946.

[10]   Coyne J. A. and Orr H. A., The evolutionary genetics of speciation, Phil. Trans. R. Soc. B (London) 353 (1998) pp. 287-305.

[11]   Darwin C. R., The Origin of Species by Means of Natural Selection (John Murray, London, 1859).

[12]   Darwin, C. R., The Descent of Man (John Murray, London, 1874) pp. 926.

[13]   Dobzhansky T., Genetics and the Origin of Species (Columbia University Press, New York, 1937).

[14]   Dobzhansky T., Genetic nature of species differences, Am. Nat. 71 (1937) pp. 404-420.

[15]   Forsdyke D. R., Two levels of information in DNA. Relationship of Romanes' "intrinsic" peculiarity of the reproductive system, and Bateson's "residue," to the species-dependent component of the base composition, (C+G)%, J. Theor. Biol. 201 (1999) pp. 47-61. (Click Here)

[16]   Forsdyke D. R., Haldane's rule: hybrid sterility affects the heterogametic sex first because sexual differentiation in on the path to species differentiation, J. Theor. Biol. 204 (2000) pp. 443-452. (Click Here)

[17]   Forsdyke D. R., The Origin of Species, Revisited (McGill-Queen's University Press, Montreal, 2001). (Click Here)

[18]   Forsdyke D. R., William Bateson. Nature Encyclopedia of Life Sciences 3 (2002) pp. 115-117. (Click Here)

[19]   Forsdyke D. R. and Mortimer J. R., Chargaff's legacy, Gene 261 (2000) pp. 127-137. (Click Here)

[20]    Goldschmidt R., The Material Basis of Evolution (Yale University Press, New Haven, 1940).

[21]    Gould S. J., Is a new and general theory of evolution emerging?, Paleobiology 6 (1980) pp. 119-130.

[22]   Guyer M. F., Hybridism and the germ cell, Bull. Univ. Cincinnati 21 (1902) pp. 1-20. (Click Here)

[23]    Haldane J. B. S., The theory of evolution, before and after Bateson,  J. Genet. 56 (1957) pp. 11-27.

[24]    Johnson N. A., Sixty years after "isolating mechanisms, evolution and temperature": Muller's legacy. Genetics 161 (2002) pp. 939-944.

[25]   King M., Species Evolution. The Role of Chromosome Change (Cambridge University Press, Cambridge, 1993).

[26]    Kliman R. M., Rogers B.T. and Noor M. A. F., Differences in (G+C) content between species: a commentary on Forsdyke's "chromosomal viewpoint" of speciation. J. Theor. Biol. 209 (2001) pp. 131-140.

[27]   Lynch M. and Force A., Gene duplication and the origin of interspecific genomic incompatability. Am. Nat. 156 (2000) pp. 590-605.

[28]   Muller, H. J., Bearing of the Drosophila work on systematics. In The New Systematics, ed. by Huxley J. S. (Clarendon Press, Oxford, 1940) pp. 185-268.

[29]    Navarro A. and Barton N. H., Chromosomal speciation and molecular divergence - accelerated evolution in rearranged chromosomes, Science 300 (2003) pp. 321-324.

[30]   Olby R., The Path to the Double Helix (University of Washington Press, Seattle, 1974).

[31]    Orr H. A., Dobzhansky, Bateson, and the genetics of speciation, Genetics 144 (1996) pp. 1331-1335.

[32]    Page A. W. and Orr-Weaver T. L., Stopping and starting the meiotic cycle, Curr. Opin. Genet. Devel. 7 (1996) pp. 23-31.

[33]   Presgraves D. C., Patterns of post-zygotic isolation in lepidoptera, Evolution 56 (2002) pp. 1168-1183.

[34]    Schlichting C. D. and Pigiucci M., Phenotypic Evolution: A Reaction Norm Perspective (Sinauer Assoc., Sunderland, MA, 1998) pp. 40.

[35]    Templeton A. R. (1982) Why read Goldschmidt?, Paleobiology 8 (1982) pp. 474-481.

[36]    White M. J. D., Modes of Speciation (Freeman, San Francisco, 1978).

Return to: Bateson and Saunders 1902 (Click Here)

Go to: Heredity and Variation 1909 (Click Here)

Go to: Chromosomal Speciation: A Reply (2004) (Click Here)

Go to: Video Lecture on Bateson & Romanes (Click Here)

Return to: Evolution Index (Click Here)

Return to: Homepage (Click Here)

This page was established in 2003 and last edited 07 Nov 2020 by Donald Forsdyke