POPULATION GENETIC ANALYSIS OF A NARROW HYBRID ZONE BETWEEN BOMBINA BOMBINA L. AND B. VARIEGATA L. (ANURA; DISCOGLOSSIDAE) IN POLAND

J.M. Szymura1, J.B. Mitton2 and W.P. Hall2,3

1 Department of Comparative Anatomy, Jagellonian University 3 Karasia 6, 30-060 KrakÛw, Poland; 2 Department of EPO Biology, University of Colorado, Boulder, Colo. 80309, USA; 3Department of Genetics, University of Melbourne, Parkville, Vic. 3052, Australia.

SUMMARY (1979)

Bombina bombina (the fire-bellied toad) and B. variegata (the yellow-bellied toad) hybridize freely wherever their ranges meet parapatrically. In southern Poland, where the species have probably hybridized at least since the last deglaciation, 5 unlinked enzyme loci provide markers which are completely diagnostic beyond the hybrid zone. The state of genetic equilibrium was studied for single- and multi-locus genotypes in 1260 individuals in 26 samples from 23 sites in a 40 km x 30 km transect across the zone west of KrakÛw. Nine of the 10 marker alleles are neutral in fitness compared to their alternatives in the foreign genome. The 10th allele seems to be superior in the foreign genome. Backcrossing and introgression occur freely, with only slight indications of a deficiency of heterozygotes; but paradoxically, introgressed foreign alleles have not escaped the narrow confines of the hybrid zone. These observations are used in three ways:

  1. All hybrid zone models are tested against this phenomenal data base. All but the hybrid sink model of Hall (1977) are rejected as plausible explanations for our seemingly paradoxical set of observations. The sink model assumes that hybrids and/or recombinants have reduced fitnesses compared to pure parental genomes. The reduced fitness of "hybrids" draws a net migration of "pures" towards the sink provided by unfit hybrid gametes and zygotes which fail to reproduce the genes that they carry. With suitable fitness and vagility parameters, all of the genes entering the sink will be trapped by the net influx, to be consumed by the unfit hybrids.

  2. The Bombina case and others like it are especially important for testing the limits and limitations of species concepts. Both the observations and the hybrid sink theory show that two parapatric populations may remain completely isolated genetically, and thus good evolutionary species, even when they hybridize freely and several generations of backcrossing and "introgression" occur. Because any gene will have only a limited lifetime in a sink unless it completely protects its carriers from mismatings, the sink effect should also prevent antihybridization selection in the hybrid zone from favoring the evolution of premating isolation mechanisms based on multiple loci.

  3. The paper is intended to be a paradigmatic example of how to assign a mechanism to a particular case of an evolutionary process when the process cannot easily be manipulated in the laboratory.

Note (2004, revised 2006)

This paper was nearly complete when it was packed for my move from Australia to the US in August/September 1979. Working conditions to unpack my research materials and library in the US to finalize the work proved to be impossible. Due to demographic circumstances of declining university enrollments and my lack of career progress, it became inescapably clear to me that I had reached a professional dead end as an evolutionary biologist. With very limited financial resources and essentially no academic support, I moved back to Australia where I had better academic contacts. By the time my library and papers appeared after spending six months on a wharf in Singapore, I was physically and psychologically unable to do any further work in biology. The present MS was not unpacked until I commenced building my personal Web pages in 2004 after more than 20 years back in Australia working in computer literacy training, software documentation, banking and the defence industry. 

The MS, as scanned from the surviving typescript, was complete pending a few responses from Mitton and Szymura to questions I set into the typescript that were not expected to change the conclusions in any way. These author's questions are highlighted in the electronic version to separate them from the running text of the MS by setting them in a different colored font. The MS as scanned, incorporates a few minor additions and corrections received from Szymura after the typescript was completed. In a very few instances where the meaning of the writing may be unclear I have added or changed a word or phrase. All such changes from the 1979 MS are set in normal type enclosed in square brackets.

For reasons I can understand and sympathize with; when queried, Mitton requested that the manuscript not be posted on my web site. Szymura did not respond to my queries. However, I will make the full MS as scanned available on written request if there is a genuinely scholarly interest in the development of the hybrid sink model where the zone of contact may serve as barriers to gene flow between hybridizing populations.

Szymura's recent papers on Bombina's marvelous natural experiment in evolutionary biology are listed on Department of Comparative Anatomy, Institute of Zoology, Jagiellonian University web site. The Contents below, outline what would have been a classic paper in evolutionary biology.

 

TABLE OF CONTENTS

INTRODUCTION

    THE BOMBINA OF EUROPE (Szymura and Hall)

       THE BIOLOGY OF BOMBINA

            1. Relationships
            2. Differentiation of the Species
            3. Geographic Distribution
            4. Ecology
                a. Breeding ponds
                b. Overwintering
                c. diet
                d. reproduction and longevity
                e. vagility
            5. Cytogenetics

        HYBRIDIZATION BETWEEN B. BOMBINA AND B. VARIEGATA

            1. Morphometric Studies
            2. Biochemical Studies
            3. Morphology and Electrophoresis

MATERIALS, METHODS AND THE ENVIRONMENTAL BACKGROUND

    SAMPLING GENES AND VAGILITY (Szymura)

    POPULATIONS SAMPLED ELECTROPHORETICALLY IN THE KRAK²W TRANSECT (Szymura)

    ENVIRONMENTS, ECOLOGY AND DISTRIBUTION OF BOMBINA IN THE AREA OF THE TRANSECT (Szymura)

    ANALYSIS OF THE DATA (Mitton, Hall and Szymura)

        Geographic variation

        Genetic equilibria

RESULTS (Mitton, Szymura and Hall)

    GEOGRAPHIC VARIATION IN GENE FREQUENCY

    DIFFERENTIAL INTROGRESSION

    GENETIC DISEQUILIBRIA

        Hardy-Weinberg Tests

        Average Frequency of Heterozygotes

        Multilocus Genotypes

DISCUSSION (Hall, Szymura and Mitton)

    PARADOXES 

    EXPLANATIONS TO RESOLVE THE PARADOXES

        The Alternatives

            Historical accident versus dynamic steady state models
            Barriers to dispersal versus selection and behavior
            Behavior versus selection
            Selection models
                1. Selection changing in direction along an environmental or genetic gradient
                2. Hybrid advantage in a hybrid habitat
                3a. Hybrid disadvantage
                3b. Hybrid disadvantage causing a "hybrid sink"

        Building an Explanation for the Paradoxes

            Choosing a selection model
            Qualifying the hybrid sink model
            The unknown and uncontrolled role of topography in a gravitational sink for gene flow

SUMMARY AND CONCLUSIONS

    THE HYBRID ZONE AS A SINK FOR GENE FLOW

    HYBRID SINKS AND THE SPECIES PROBLEM

REFERENCES

APPENDICES

    APPENDIX 1. LDH-1

    APPENDIX 2. MDH-1

    APPENDIX 3. CK

    APPENDIX 4. AK

    APPENDIX 5. GPI

    APPENDIX 6. Frequency of B. bombina alleles in 26 populations of Bombina from the transect west of KrakÛw, Poland

    APPENDIX 7. Indices of heterozygosity and hybridity for 26 samples of Bombina from the transect west of KrakÛw, Poland, and their inferred distances from the center of the hybrid zone

TABLES

    TABLE 1. Average gene frequencies and heterogeneity of frequencies in 21 population samples from the Bombina hybrid zone west of KrakÛw

    TABLE 2. Average gene frequencies and heterogeneities of frequencies between population samples from different years, etc.

    TABLE 3. Tests of heterogeneity of variances of allelic frequencies for 5 polymorphic loci across 21 hybrid populations of Bombina

    TABLE 4. Principal components analysis of genetic variation in 26 Bombina populations from the transact west of KrakÛw, Poland.

    TABLE 5. Sign tests of deviations from Hardy-Weinberg expectations (D) in populations from central areas of the transect through the Bombina hybrid zone west of KrakÛw, Poland

    TABLE 8? Improbable genotypes in population samples from the Bombina hybrid zone west of KrakÛw

FIGURES

    Figure 1. Distribution of Bombina in Central Europe

    Figure 2. Geography of the KrakÛw transect.

    Figure 3. Tyniec sample sites

    Figure 4. Frequency of B. bombina alleles, f(b) versus geography.

    Figure 5. Frequency of B. bombina alleles versus distance from center of hybrid zone.

    Figure 6. Frequency of B. bombina alleles at single loci.

    Figure 7a. Heterozygosity in B. variegata-like populations versus average frequency of B. bombina alleles.

    Figure 7b. Heterozygosity on B. bombina-like populations versus average frequency of B. bombina alleles.