THE EXPECTED FIXATION RATE OF CHROMOSOMAL INVERSIONS.

Abstract

Most types of major chromosomal rearrangements substantially reduce the fertility of heterozygous carriers due to the segregation of aneuploid gametes, although rearrangement homozygotes usually segregate regularly and have nearly normal fitness (Burnham, 1962; White, 1973). Exceptions to the rule of strong heterozygote disadvantage occur for some kinds of rearrangements, such as fusions (or fissions) between telocentric (or acrocentric) chromosomes of similar length, and small inversions, and for paracentric inversions in Diptera where special mechanisms allow heterozygotes to circumvent the production of aneuploid gametes (White, 1973). Nevertheless, major chromosomal rearrangements which appear to have caused considerable heterozygote disadvantage, such as large inversions and asymmetric fusions, have been fixed in evolution at the rate of roughly 10-6 to 10-7 per generation in vertebrate lineages, and about an order of magnitude more slowly on average in invertebrates (Bush et al., 1977; Lande, 1979; Imai et al., 1983), so that closely related species often differ karyotypically by such rearrangements (White, 1973, 1978). These fixation rates are orders of magnitude slower than the spontaneous rearrangement rates, which are roughly 10-3 to 10-4 per gamete per generation in various species (White, 1978 Ch. 6; Lande, 1979). Several theories have been proposed to explain the fixation of major chromosomal rearrangments, including random genetic drift in small populations (Wright, 1941; Bengtsson and Bodmer, 1976; Lande, 1979), or a deterministic advantage such as natural selection caused by breakpoint or position effects, or meiotic drive, or some combination of factors (White, 1973, 1978; Bickham and Baker, 1979; Bengtsson, 1980; Bush, 1981; Hedrick, 1981; Walsh, 1982). Because many rearrangements, especially inversions, effectively suppress recombination around the breakpoints, and within the inverted regions, they may also be selected by association with a favorable combination of linked alleles (Sturtevant and Mather, 1938; Nei et al., 1967; Ohta and Kojima, 1968; Van Valen and Levins, 1968; Dobzhansky, 1970; Charlesworth and Charlesworth, 1973; Charlesworth, 1974). Major cytologically detectable chromosomal rearrangements are capable of producing the same range of phenotypic effects as other cytologically indetectable mutations (Lindsley and Grell, 1968), in addition to mosaic phenotypes (variegated position effects) which appear to be of little evolutionary significance (Muller, 1 956). But in most eukaryotic species, chromosomal rearrangements usually do not produce noticeable phenotypic effects. (other than reducing heterozygote fertility), and are not known to confer any deterministic advantage; hence they may have been fixed by random genetic drift in small populations (Lande, 1979). In many taxa karyotypic differences exist between sibling species which are virtually identical morphologically (White, 1973). Even in Drosophila, abundant polymorphisms for paracentric inversions do not influence the morphological characters usually employed by taxonomists (Spieth and Heed, 1972). However, White (1969, 1978 Ch. 6) has pointed out that the few rearrangements which are fixed may be a highly selected subset of those which arise spontaneously, thus