5-azacytidine-treated human fibroblasts exhibit a significant decrease in the frequencies of Barr body+ cells as compared to nontreated cultures. This presumably indicates that 5-azacytidine can induce a change in the degree of condensation of the Barr body. It is suggested that the state of chromatin condensation of the Barr body may be related to the reactivation process by 5-azacytidine of gene loci in the inactive X.

The T/t–H-2 region on mouse chromosome 17 is known from complex natural variants (‘t-haplotypes’) to contain numerous genes, including some affecting the immune system and the development of the embryo. Rapid progress in the isolation of recombinant DNA clones for this 50 megabasepair region is generating the material for its complete molecular anatomy. A crucial step in revealing the biological functions controlled by the region is to obtain mutants in which genes are inactivated individually. We have used a pair of inbred mouse strains and a series of classical breeding schemes that permit the detection of recessive lethal and detrimental mutations in the T/t–H-2 region.

In this initial phase of our study, 280 gametes mutagenized in the male germ line by ethylnitrosourea (ENU) have yielded eleven independent pre-natal recessive lethal mutations. Four have been mapped against T mutations and have been shown to complement one another in all pairwise combinations.

Mating competition experiments with D. melanogaster were performed in light and darkness under conditions mimicking those under which artificial populations show a stable polymorphism for the ebony gene (e). In both light and darkness a minority male mating advantage was found for e/ + males in competition with wild-type males, but in light only when ebony was in the majority. Homozygous wild-type males showed minority advantage only in light and ebony males only in darkness.

Seven strains of guinea pig (Cavia porcellus) with different degrees of inbreeding were investigated for genetic variability by starch gel electrophoresis. Variants for six out of 31 loci (Gpd-1, Aat-1, Ak-2, Pgm-1, Ada and Gpi) are reported. The data suggest that the strains all come from a small founding population and therefore are genetically more homogenous than the visible characteristics imply. Differences between isoenzyme data and other genetic characters are described and possible underlying genetic mechanisms are discussed.

A temporal survey of the alcohol dehydrogenase locus in a natural population of Drosophila melanogaster from the Canary Islands has revealed the existence of an electrophoretic Fast thermostable variant, mapped within 0.8 units of the structural Adh locus, and which persists over a 6-year period with a frequency ranging from 0.8 to 0.5%.

The maintenance of polygenic variation through a balance between mutation and stabilizing selection can be approximated in two ways. In the ‘Gaussian’ approximation, a normal distribution of allelic effects is assumed at each locus. In the ‘House of Cards’ approximation, the effect of new mutations is assumed to be large compared with the spread of the existing distribution. These approximations were developed to describe models where alleles may have a continuous range of effects. However, previous analyses of models with only two alleles have predicted an equilibrium variance equal to that given by the ‘House of Cards’ approximation. These analyses of biallelic models have assumed that, at equilibrium, the population mean is at the optimum. Here, it is shown that many stable equilibria may coexist, each giving a slight deviation from the optimum. Though the variance is given by the ‘House of Cards’ approximation when the mean is at the optimum, it increases towards a value of the same order as that given by the ‘Gaussian’ approximation when the mean deviates from the optimum. Thus, the equilibrium variance cannot be predicted by any simple model, but depends on the previous history of the population.

Forty-one third chromosomes extracted from a natural population of Drosophila melanogaster were assessed for net fitness and for the quantitative characters viability, net fertility, female productivity, male weight, abdominal bristle number, and sternopleural bristle number. Net homozygous and heterozygous fitness of the third chromosomes was estimated by competition against a marked balancer third chromosome. Average fitness of the homozygous lines relative to wild-type heterozygotes was 0·13, indicating substantial inbreeding depression for net fitness. All significant correlations of quantitative characters with fitness and with each other were high and positive. Homozygous fitness is strongly correlated with net fertility, viability, and female productivity, moderately associated with male weight, and not significantly associated with bristle traits. The combination of metric traits which best predicts homozygous fitness is the simple multiple of viability and female productivity. Heterozygous fitness is not correlated with homozygous fitness; furthermore, the relative contribution of metric traits to fitness in a heterozygous population is likely to be different from that deduced from homozygous lines. These observations are consistent with a model of genetic variation for fitness in natural populations caused by segregation of rare deleterious recessive alleles.

Data from foetal and weanling studies suggest that Oxford race metacentrics kq and no, but not pr, are transmitted preferentially from Robertsonian heterozygote common shrews collected from the Oxford area. These results suggest a possible mechanism for the maintenance of the Robertsonian polymorphism in that area.

An analysis based on a model that is different from the traditional Fisher's model for quantitative characters under assortative mating reveals that the genotypic offspring–midparent regression can be affected by assortative mating of parents. It is demonstrated that the prediction that mating parents assortatively introduces only a negligible bias in the estimated coefficient of linear offspring-midparent regression is limited to Fisher's model and cannot be generalized.

Recombination is hard to understand in darwinian terms when the process is identified with the production of crossover chromosomes. As an alternative explanation I propose instead that biased conversion is the primary function of meiotic recombination. In particular I show that a conversion process directed against the most common type of genetic damage can substantially reduce the mutational load, even if the conversion force is weak and if the conversion process occasionally creates new mutations.