Asynapsis and mutability in Drosophila melanogaster.

I N Drosophila mekanogaster heterozygosity for simple or multiple autosomal inversions results in increased mutability of the structurally heterozygous chromosomes (THOMPSON 1959, 1960). This has been shown in studies of both spontaneous and induced mutations. Use of the “brood pattern” method ( AUERBACH 1954) following irradiation established that the increase in mutability is limited to meiotic or immediately premeiotic stages. The extreme sensitivity of inversion heterozygotes to X-irradiation during these stages suggests a synergistic interaction of irradiation and the genetic constitution (THOMPSON 1960), and it seemed likely that localized asynapsis at the points of change in linear homology might be effective in causing the failure of repair of an initially high level of mutational events in these regions. This idea is borne out by the distributions of lethals obtained from lines heterozygous for different inversions. There has been a strikingly preferential occurrence of mutations adjacent to the breakpoint loci of certain of the corresponding inversion (s) . An alternative to this role of asynapsis which deserves consideration is based on the observation (NOVITSKI 1946; ROTHFELS and FAIRLIE 1957) that in natural populations of organisms where chromosomal polymorphism prevails, new inversion breaks appear to occur preferentially at or near the loci of existing inversion breaks. This phenomenon has been attributed to breakage by stress where the changes in homology cause conflicting attractions, and to the proximity of newly broken ends in those regions (NOVITSKI 1946). Such an effect of synaptic stress or torsion at the genic level might cause preferential mutations in the regions corresponding to existing breaks, as in the present case. On the other hand, it is possible that in the cases examined by these authors the mechanism involved is a failure of break restitution in unsynapsed regions rather than, or in conjunction with, a role of localized stress in raising the level of break formation. It would be desirable, then, to evaluate the relative importance of asynapsis and pairing stress in the increase of mutations. These variables are difficult to isolate. There is no apparent means of creating synaptic tensions without concomitant asynapsis, since tension implies the existence of a disruption of linear homologies that will necessarily result in some nonpairing of homologues. It has seemed possible, however, to achieve total or nearly complete asynapsis of an entire chromosome pair and thus to eliminate pairing tension as a factor. This