EXTENSIVE CHROMOSOMAL DIVERGENCE WITHIN A SINGLE RIVER BASIN IN THE GOODEID FISH, ILYODON FURCIDENS

Extensive variation in the number of metacentric chromosomes exists among populations of the viviparous goodeid fish, Ilyodon furcidens, in the Río Coahuayana basin of south central Mexico (states of Colima and Jalisco). The variation can be divided, somewhat arbitrarily, into four "cytotypes" with 0-2, 0-4, 6 and 10-16 metacentrics. Of these, the first, shared with the closely adjacent Río Armería and similar to other species of Ilyodon, is probably ancestral. The wholly non-Robertsonian nature of the variation and its extent appear to be unprecedented among teleosts, but its uniqueness is difficult to evaulate because fish chromosome data in general may be biased toward both monomorphism and Robertsonian variation. Variation is evident with all cytotypes but has been well characterized for only a single population of the 0-4M cytotype. That population, unlike most of the others, consists of two interbreeding morphs which differ in mouth width. The variation is heterogeneously distributed between the morphs; the significance of this observation is not yet clear. The distribution of the cytotypes is approximately clinal with respect to the number of metacentric chromosomes. Although the cline may be a direct response to some gradient in selection intensity, the possibility that it is the result of secondary contact of previously isolated populations, fostered by tributary transfer, is real. Allozyme comparisons reveal minimal genic divergence among the cytotypes. There are no fixed allelic differences, and the average unbiased genetic distance between the two extreme cytotypes is 0.042. Gene diversity analysis indicates that an average of less than 3% of the total variation (H_T = 0.072) is partitioned among cytotypes; about 24% is partitioned among populations within cytotypes. Genic and chromosomal divergence in Ilyodon are clearly uncoupled. Laboratory F₁ , backcross, and F₂ intercytotype hybrids are fully viable, and are indistinguishable in fertility from our stocks derived from single populations. F₃ intercytotype hybrids are also fully viable but have not yet been tested for fertility. This suggests that, during the course of chromosomal evolution, single rearrangement heterozygotes were not appreciably negatively heterotic, even though the rearrangements are probably pericentric inversions. The combined data suggest that the chromosome rearrangements, even in multiple form, do not function as significant isolating mechanisms. Chromosomal evolution in Ilyodon, though quite marked, has apparently not fostered speciation.