A study of α-amylase isozyme patterns from gibberellin-induced endosperms of wheat-alien genotypes (amphiploid, addition and substitution lines) resolved by flat-bed isoelectric focusing identified homoeoloci for α-Amy-1 (malt α-AMY-1 genes) on chromosomes 6H of Hordeum vulgare, 6RL of Secale cereale, 6Rm of S. montanum and 6E of Agropyron elongatum. Homoeoloci for α-Amy-2 (green α-AMY-2 genes) were identified on chromosomes 7HchL of Hordeum chilense, 7RL of Secale cereale, 7Sb of Aegilops bicornis, 7U of Ae. umbellulata and 7EL of Agropyron elongatum. Analysis of mature grain β-amylase identified β-Amy-1 loci on chromosomes 4H of H. vulgare, 4Hch of H. chilense, 4S1 of Ae. sharonensis and Ae. longissima and β-Amy-2 loci on chromosomes 5RL of S. cereale and 5U of Ae. umbellulata. These gene locations provide further evidence for the homoeology of the alien chromosomes with wheat and for the conservation of gene synteny among wheat and its relatives.

Wild-type Caenorhabditis elegans fails to reach adulthood if L1 larvae are incubated in the presence of 30 mM or greater concentrations of caffeine. Eleven mutants have been isolated in which caffeine has a less pronounced effect on development. The mutations are recessive, define two genes, and have been mapped. The mechanism(s) of resistance is unknown.

The glue proteins are products of a developmentally regulated gene family. These genes are transcriptionally active during the third larval instar and code for the major protein products of salivary glands. The activity of several of the genes can be visualized as intermoult puffs in the polytene salivary gland chromosomes. The amount of one of these proteins, P5, varies widely among wild-type strains. We have used biochemical and genetic methods to investigate the source of this variation. The results of in vitro translation of salivary gland RNA suggest that the variation occurs pretranslationally. Genetic mapping experiments showed that sites on several chromosomes can modulate the amount of P5, but that one site on the third chromosome determines the absence and presence of this protein. We have mapped this glue protein gene, called GP5, to the interval between bx (3–58·8) and sr (3–62·0) which also includes the intermoult puff at 90BC. We discuss the relationship between P5 and the glue protein gene Sgs-5 which is also located at 90BC.

In order to increase our understanding of the evolutionary dynamics of transposable genetic elements we have studied the chromosal location of copies of 2 element families in 20 X chromosomes extracted from a natural population of Drosophila melanogaster from Spain. The I element was localized at a total of 64 chromosomal sites and copia at 45 sites in this sample with a mean copy number of 3·2 and 2·3 elements/chromosome respectively. Both elements were highly variable in location, with no site reaching a higher frequency than 4/20 in either case. Comparisons with other data sets suggest that insertion frequencies can be used to detect population structuring.

The ability of restricted selection indices to prevent genetic change in a restricted trait over several generations of selection was studied using deterministic computer models. Four loci, two affecting each trait independently, and two pleiotropic loci, one affecting each trait in the same direction, and one with opposite effects, were modelled. In general, continued effectiveness of the restriction was achieved only when the restricted trait was affected by only one locus. In some conditions (equal gene frequencies), an independent locus and one pleiotropic locus affecting the restricted trait allowed maintenance of the restriction. The results suggest that long-term restriction may be very difficult without re-estimation of parameters.

The accumulation of beneficial and harmful mutations in a genome is studied by using analytical methods as well as computer simulation for different modes of reproduction. The modes of reproduction examined are biparental (bisexual, hermaphroditic), uniparental (selfing, automictic, asexual) and mixed (partial selfing, mixture of hermaphroditism and parthenogenesis). It is shown that the rates of accumulation of both beneficial and harmful mutations with weak selection depend on the within-population variance of the number of mutant genes per genome. Analytical formulae for this variance are derived for neutral mutant genes for hermaphroditic, selfing and asexual populations; the neutral variance is largest in a selfing population and smallest in an asexual population. Directional selection reduces the population variance in most cases, whereas recombination partially restores the reduced variance. Therefore, biparental organisms accumulate beneficial mutations at the highest rate and harmful mutations at the lowest rate. Selfing organisms are intermediate between biparental and asexual organisms. Even a limited amount of outcrossing in largely selfing and parthenogenetic organisms markedly affects the accumulation rates. The accumulation of mutations is likely to affect the mean population fitness only in long-term evolution.

Classical population genetic models show that disruptive selection in a spatially variable environment can maintain genetic variation. We present quantitative genetic models for the effects of disruptive selection between environments on the genetic covariance structure of a polygenic trait. Our models suggest that disruptive selection usually does not alter the equilibrium genetic variance, although transient changes are predicted. We view a quantitative character as a set of character states, each expressed in one environment. The genetic correlation between character states expressed in different environments strongly affects the evolution of the genetic variability. (1) If the genetic correlation between character states is not ± 1, then the mean phenotype expressed in each environment will eventually attain the optimum value for that environment; this is the evolution of phenotypic plasticity (Via & Lande, 1985). At the joint phenotypic optimum, there is no disruptive selection between environments and thus no increase in the equilibrium genetic variability over that maintained by a balance between mutation and stabilizing selection within each environment. (2) If, however, the genetic correlation between character states is ± 1, the mean phenotype will not evolve to the joint phenotypic optimum and a persistent force of disruptive selection between environments will increase the equilibrium genetic variance. (3) Numerical analyses of the dynamic equations indicate that the mean phenotype can usually be perturbed several phenotypic standard deviations from the optimum without producing transient changes of more than a few per cent in the genetic variances or correlations. It may thus be reasonable to assume a roughly constant covariance structure during phenotypic evolution unless genetic correlations among character states are extremely high or populations are frequently perturbed. (4) Transient changes in the genetic correlations between character states resulting from disruptive selection act to constrain the evolution of the mean phenotype rather than to facilitate it.

The maintenance of polygenic variability by a balance between mutation and stabilizing selection has been analysed using two approximations: the ‘Gaussian’ and the ‘house of cards’. These lead to qualitatively different relationships between the equilibrium genetic variance and the parameters describing selection and mutation. Here we generalize these approximations to describe the dynamics of genetic means and variances under arbitrary patterns of selection and mutation. We incorporate genetic drift into the same mathematical framework.

The effects of frequency-independent selection and genetic drift can be determined from the gradient of log mean fitness and a covariance matrix that depends on genotype frequencies. These equations describe an ‘adaptive landscape’, with a natural metric of genetic distance set by the covariance matrix. From this representation we can change coordinates to derive equations describing the dynamics of an additive polygenic character in terms of the moments (means, variances, …) of allelic effects at individual loci. Only under certain simplifying conditions, such as those derived from the Gaussian and house-of-cards approximations, do these general recursions lead to tractable equations for the first few phenotypic moments. The alternative approximations differ in the constraints they impose on the distributions of allelic effects at individual loci. The Gaussian-based prediction that evolution of the phenotypic mean does not change the genetic variance is shown to be a consequence of the assumption that the allelic distributions are never skewed. We present both analytical and numerical results delimiting the parameter values consistent with our approximations.