The frequency and distribution of chiasmata may vary between species, between genotypes, between cells and even between bivalents within cells, these differences may be due to genetic or environmental causes (Jones 1987). One restriction on the number of chiasmata per bivalent is chiasma interference. With very small bivalents this can limit the number of chiasmata to one. Excepting very small chromosomes, Mather (1937) showed that the number of chiasmata per bivalent was directly proportional to chromosome length.In six Lolium species, Rees and Narayan (1988) found a correlation between DNA amount and chiasma frequency. Vincent and Jones (1993), in autotriploid Scilla autumnalis, showed a trend where the longer chromosome sets had more pairing partner exchanges than the shorter chromosomes at prophase I. However, Wang and Berdahl (1990) when comparing Triticeae species found that trivalent frequency is not related to chromosome length. However, trivalent frequency is partly dependent on chiasma numbers (Thomas 1995).Analysis of synaptonemal complex formation in two aneuploid genotypes of autotriploid L. multiflorum revealed that 94% of homologous groups formed trivalents at prophase I. The trivalents had on average two pairing partner exchanges at pachytene and the two genotypes respectively had 3.1 and 2.7 chiasmata per homologous group at metaphase I. This was sufficient to ensure that in the two genotypes respectively, 88% and 86% of the trivalents formed during prophase I were retained to metaphase I. The rate of retention indicated that most of the chiasmata occurred in separate SC regions (Thomas and Thomas 1994). L. temulentum has 50% more DNA per nucleus than L. multiflorum and correspondingly longer chromosomes (Hutchinson et al. 1979) and a higher chiasma frequency at the diploid level (Rees and Narayan 1988). Thomas (1994) analyzed the frequencies of chiasmata, paired arms and chiasmata per paired arm in diploid L. multiflorum and L. temulentum. Chiasma frequency and chiasma distribution had separate control mechanisms so that a L. multiflorum genotype with a high chiasma frequency had a lower frequency of paired arms than a genotype of L. multiflorum with a low chiasma frequency. However, L. temulentum with its larger chromosomes, had a higher chiasma frequency and a higher frequency of paired arms than the L. multiflorum genotypes. The present report describes configurations at metaphase I and the distribution of chiasmata in autotriploid cytotypes of these two species.
Materials and methodsDiploid and tetraploid cytotypes of L. multiflorum ssp. westerwoldicum Bb2074 and L. temulentum Ba3081 were crossed to produce the autotriploids. The autotriploids of L. multforum were produced by enclosing inflorescences of the diploid and tetraploid cytotypes in a pollination bag prior to anther dehiscence. Immature embryos were removed 18 days after pollination and cultured on a modified Gamborg B5 medium (Gamborg et al. 1968) with 3% sucrose but without 2, 4D and kinetin. The autotriploid L. temulentum was s...