Cytogenetic Analyses of Autopolyploids: Models and Methods for Triploids to Octoploids

Abstract

Methods are presented for determining the frequencies and numbers of various meiotic configurations expected in autopolyploids. This allows one to test polyploids of unknown origin for agreement with expected meiotic configurations. Rejection of the autoploid hypothesis may indicate the presence of Ph‐like genes or some type of alloploid. The models consider mean chiasma frequencies of 2, 3, and 4 per bivalent for triploids and tetraploids and 2 per bivalent for pentaploids, hexaploids, heptaploids, and octoploids. Literature data for a known autotriploid, autotetraploids, allotetraploids, and allopentaploids were tested against expectations of the models. There was generally good agreement between number of observed autoploid meiotic configuration and those expected in the models.

Early adjustment of patterns of metaphase association in the evolution of a polyploid species

“…Table 2 gives the expected frequencies of chromosomal associations using the models put forward by Jackson & Casey (1982) which assumes autopolyploid behaviour. The values in the table are based on the chiasma frequencies of the Welsh, the synthesised and the Edinburgh populations of S. cambrensis.…”

The control of chiasma frequeney within a polyploid series in the genus Senecio (Compositae)

“…Chromosomal configurations were classified as olV (ring quadrivalents), clV (chain quadrivalents), III (trivalents), oli (ring bivalents), clI (chain bivalents), I (univalents) following the usage of Jackson and Casey (1982).…”

Further evidence for the presence of meiotic pairing control genes in Alopecurus L. (Gramineae)

“…Until recently methods have not been available to predict the frequency of meiotic configurations in autopolyploids, and the classification of plants as autoploids or segmental alloploids has been highly subjective. Therefore the publication of models and formulae by Jackson and Casey (1982) and Jackson and Hauber (1982), which can be used to predict the frequency of meiotic configurations in autopolyploids are of great value. The basis of these models is the assumption that chromosome pairing(synapsis) in normal autopolyploids is a random process but that the distribution of chiasmata is, in most situations, non-random.…”

“…On the basis of random distribution of 0 to 2 chiasmata per bivalent, univalents, like circle and chain bivalents, should occur at frequencies predicted by the binomial (p + q)" where p = presence of chiasmata, q = absence of chiasmata and n = maximum number of chiasmata per configuration (= bivalent). However, from an examination of ourownandotherdataitisclearthatnaturalgenetic systems do not permit univalent formation and therefore their occurrence is sporadic and their frequency cannot be predicted (Jackson, 1982). It is logical to extrapolate that univalent formation in normal autopolyploids is also unpredictable, and to overcome this problem Jackson and Hauber (1982) have devised correction factors for the original binomial models of Jackson and Casey (1982) which eliminate the univalent class.…”

“…However, from an examination of ourownandotherdataitisclearthatnaturalgenetic systems do not permit univalent formation and therefore their occurrence is sporadic and their frequency cannot be predicted (Jackson, 1982). It is logical to extrapolate that univalent formation in normal autopolyploids is also unpredictable, and to overcome this problem Jackson and Hauber (1982) have devised correction factors for the original binomial models of Jackson and Casey (1982) which eliminate the univalent class.…”