MOST OF THE HOMOEOLOGOUS PAIRING AT CHIASMATIC METAPHASE I IN WHEAT-RYE HYBRIDS IS NOT

The use of telomeric C-bands in wheat-rye hybrids has made it possible to distinguish three types of wheat-wheat ( IBL) and wheat-rye associations (a, endto-end extremely distal; b, end-to-ed distal; and c, interstitial) between homoeologous chromosomes at different metaphase I stages (early, middle and late) and also to estimate the actual recombination frequencies for such associations at anaphase 1. There was a decrease of the a and b association frequencies during the different metaphase 1 stages, whereas the c type remained without variation in all stages. A good tit between the frequencies of c associations at metaphase I and the number of recombinant chromosomes at anaphase I, assuming a maximum of one chiasma per bond, was found; however, there was no correspondence between metaphase I and anaphase I data when all associations (a + b + c) were considered. In addition, rye-rye homologous pairing was observed at metaphase I, but no evidence for rye-rye recombination was found at anaphase I. The results indicate that most of end-to-end (a and 6) homoeologous and nonhomologous associations are actually nonchiasmatic and are a remnant of prophase pairing. T generally has been believed that the number of associations between chroI mosome arms at metaphase I is equivalent to chiasma formation frequency (earlier meiotic stages, such as diplotene, that might give more precise indications are unreliable in higher plants). This assumption has lead many cytogeneticists and breeders to accept the idea that homoeologous pairing frequency in interspecific hybrids is a good reflection of genetic transfer between two species. However, as it has been pointed out by JONES (1978), this equating is notoriously imprecise due to the highly condensed state of the bivalents at metaphase I. In addition, careful studies comparing metaphase 1 bound arms and anaphase I recombinant chromosomes for specific chromosomal markers detected by C-banding in rye have demonstrated that the number of bound arms at metaphase I is not equivalent to the actual number of chiasmata (crossing overs) formed at first meiotic prophase, because more than one chiasma per bond could be formed in some chromosomes (GIRALDEZ and ORELLANA 19’79; ORELLANA and GIRALDEZ 198 1). Moreover, in desynaptic plants there is clear evidence that some of the bonds appearing at metaphase I are