The influence of linkage on the rates of response to continued directional selection of quantitative traits deriving from variation contributed by new mutations in finite populations is investigated. Mutant genes are assumed to have additive effects which are randomly sampled from a symmetric distribution, and to be randomly located on the chromosome. Results were obtained by Monte Carlo simulation.

The expected steady rate of response, when variability from new mutations is balanced by that lost by drift and selection, is reduced the tighter the linkage, but the reduction is small unless there are few, short chromosomes. For a given rate of new mutational variance per haploid chromosome set per generation, greater effects of linkage are obtained in large than in small populations, because more mutants segregate. The response and influence of linkage are essentially the same whether the new variance is due to many genes of small effect or few of large effect.

The variability of response between replicates and generations was investigated, and the contribution to this of new mutants or recombination of existing mutants compared. Usually most genetic variability was due to the occurrence of a new favourable mutant of large effect.

Using the 602 second chromosome lines extracted from the Ishigakijima population of Drosophila melanogaster in Japan, partial diallel cross experiments (Design II of Comstock & Robinson, 1952) were carried out, and the additive genetic variance and the dominance variance of viability were estimated. The estimated value of the additive genetic variance is 0·01754±0·00608, and the dominance variance 0·00151±0·00114, using a logarithmic scale. Since the value of the additive genetic variance is much larger than expected under mutation–selection balance although the dominance variance is compatible with it, we speculate that in the Ishigakijima population some type of balancing selection must be operating to maintain the genetic variability with respect to viability at a minority of loci. As candidates for such selection, overdominance, frequency-dependent selection, and diversifying selection are considered, and it is suggested that diversifying selection is the most probable candidate for increasing the additive genetic variance.

The effect of the P-M system of hybrid dysgenesis in Drosophila melanogaster on unequal crossing-over was studied using the Bar duplication. This system of dysgenesis had no demonstrable effect on the rate of Bar reversion. In the course of the study it was found that there was a greatly reduced rate of reversion in two homozygous inversion stocks. Further, one revertant was found which may result from unequal crossing over at the Bar locus in males.

Recombination between two mouse t–haplotypes, tw12tf and tLub-1, was investigated by screening the tailless progeny of the cross ♀ tw12tf/tLub-1 + × ♂Ttf/+tf for the segregation of tufted phenotype, tw12 and tLub-1 lethal factors, and metacentric chromosome (since tLub-l haplotype is associated with a Robertsonian fusion involving chromosomes 4 and 17). The results give a 17% estimate of the recombination frequency between centromere and tf, with tLub-l lethal factor mapping about two-thirds of the distance from centromere to tf and the tw12 lethal factor behaving as if closely linked to tf. This further extends the findings of Silver & Artzt (1981) and of Artzt, McCormick & Bennett (1982), and shows that two t–haplotypes with quite independent laboratory histories recombine at a normal level, supporting the notion that all t–haplotypes basically share the same structure.

McLaren & Monk (1982) and Cattanach et al. (1982) reported that T(X; 16)16H/X Sxr mice, in which the X chromosome bearing Sxr is the inactivated X chromosome, can develop as fertile females. By mating such females to X/Y Sxr males it has been possible to produce mice homozygous for Sxr. Two X Sxr / Y Sxr males were identified which together fathered 141 sons and 1 daughter. The single daughter proved to be XO, indicating a non-disjunctional event with neither paternal sex chromosome being transmitted. It is concluded that X Sxr / Y Sxr mice are viable and fertile, and that all their progeny, provided they receive a paternal sex chromosome, develop as males.

In extracts of mature wheat grains, 13 hexokinase isozymes were distinguished by IEF. The genes controlling the production of five isozymes were located on chromosome arms 1BS, 1DS and 3BS by nullisomic analysis. The three loci, part of two homoeoallelic series (Hk-1 and Hk-2) are designated Hk-B1, Hk-D1 and Hk-B2 respectively. Analysis of chromosome 1D short-arm terminal deletions indicated the Hk-D1 locus to be located proximally to the glucose phosphate isomerase locus, Gpi-D1 on the shortarm. Three variant HK phenotypes were distinguished amongst 55 hexaploid wheats examined. Analysis of seven Chinese Spring/Agropyron elongatum chromosome addition lines showed that Ag. elongatum isozymes were expressed in the wheat background in additions IV and V.