Myths and mechanisms of meiosis.

A comparative analysis of the meiotic secquence in a wide variety of organisms indicates there is no convincing evidence that: (1) Premeiotic pairing plays any role in the synapsis of homologues. (2) Heterochromatic association facilitates homologous pairing. (3) Chiasmata ever form within segments which are positively heteropycnotic at zygotenepachytene. (4) Localisation of chiasmata depends on prior localisation of pairing or on the occurrence of euchromatin-heterochromatin boundaries. (5) Prior association of centromeres plays any role in determing co-orientation. (6) Any form of supra-chromosomal organisation exists involving permanent association between the members of a haploid complement, and (7) Unequal progeny ratios recovered from structurally modified Drosophila complements arise as a consequence of distributive pairing.--On the other hand there is good evidence that: (1) Interlocking of bivalents can occur regularly in species with a chiasma frequency sufficiently high to regularly produce ring bivalents and in which the chiasmata are localised to the ends of the bivalent. (2) Some forms of terminal association cannot represent terminalised chiasmata. (3) U-type exchanges present at diplotene result from errors in crossing over. (4) Pairing and chiasma formation are not necessary for coorientation, and (5) at least some types of elastic constrictions present at first metaphase represent extended nucleolar organisers.