A 1·5 kilobasepair repeated DNA sequence is duplicated in direct orientation so as to flank the suppressor of forked gene in the euchromatin–heterochromatin transition region on the X chromosome of Drosophila melanogaster. These two copies are almost identical, but DNA blotting, analysis of cloned sequences and database searches show that elsewhere in the genome, homologous sequences are poorly conserved. They are often associated with other repeats, suggesting that they may belong to a scrambled and clustered middle repetitive DNA family. The sequences do not appear to be related to transposable elements and their location in different strains is conserved. In situ hybridization to metaphase chromosomes shows that homologous sequences are concentrated in the pericentric regions of the autosomes and the X chromosome. The sequences are not significantly under-represented in DNA from polytene tissue and must lie in the replicated regions of polytene chromosomes. The almost perfect conservation of the two repeats around suppressor of forked in D. melanogaster suggests they arose by duplication or gene conversion. Suppression of recombination in this chromosomal region presumably allows this unusual organization to be stably maintained. In the X-ray induced allele, suppressor of forked-L26, the sequence between the repeats, including the gene, and one copy of the repeat have been deleted.

The alcohol dehydrogenase (Adh) gene of Drosophila melanogaster is well suited to be a gene expression reporter system. Adh produces a measurable phenotype at both the enzyme and mRNA levels. We recovered a spontaneous transposable element (TE) insertion mutation near the Adh gene. The insertion is a truncated retroposable element, jockey, inserted upstream of the adult Adh enhancer region. Comparisons between the Adhjockey allele and its direct wild-type ancestral allele were made in an isogenic background (i.e. identical cis and trans factors). Differences in Adhjockey expression compared with the wild-type can be attributed solely to the presence of the jockey element. This jockey insertion results in a decrease in adult mRNA transcript levels in the Adhjockey homozygous lines relative to the wild-type counterpart and accounts for a correlated decrease in alcohol dehydrogenase (ADH) enzyme activity. The larval ADH activity levels are not detectably different.

Sexual isolation between the sibling species D. simulans and D. mauritiana is due largely to the rejection of D. simulans males by D. mauritiana females. Genetic analysis shows that genes on the X and third chromosomes contribute to the differences between males causing sexual isolation, while the Y chromosome, second chromosome and cytoplasm have no effect. These chromosome effects differ from those observed in a previous analysis of sexual isolation in hybrid females, implying that different genes cause sexual isolation in the two sexes.

The mutant doublefoot, Dbf, of the mouse arose spontaneously, and was shown to be inherited as an autosomal dominant, mapping 9–13 cM proximal to leaden, ln, on chromosome 1 and showing no recombination with the microsatellite markers D1Mit24 and D1Mit77. In heterozygotes the phenotype includes many extra toes on all four feet, and the tibia and fibula may be reduced and bowed. The head is shortened and broad and the eyes are held half-closed, and some animals develop hydrocephalus. The tail is kinked and abnormally thick, and the soles of the feet are swollen. Growth is retarded, viability is reduced, and reproduction is impaired in both sexes. Only about 30 % of males are normally fertile, and testis weights and sperm counts may be reduced, although this appears not to be the main cause of poor fertility. In females vaginal opening is delayed and oestrous cycles are irregular, although the animals appear to respond to gonadotrophic hormones. Crosses of Dbf/ + × Dbf/ + are very poorly fertile. Prenatally, Dbf/ + heterozygotes can first be recognized at 11½ days gestation by abnormally broad fore limb buds. Putative Dbf/Dbf homozygotes at 12½ days have similar limbs defects and also split face, due to failure of the maxillae to fuse in the midline. Some homozygotes and a few putative heterozygotes have cranioschisis. At 13½ days, the heads of homozygotes tend to bulge in the frontal region and a bleb of clear fluid is visible medially. At 14½ days Dbf/Dbf fetuses may have oedema and some are dead. From 15½ days onwards no live Dbf/Dbf fetuses have been found. The gene maps close to the locus of Pax3, but crossovers between Dbf and Pax3 have been found, ruling out the possibility that a gain-of-function mutation in Pax3 might be involved.

A MINQUE(l) procedure, which is minimum norm quadratic unbiased estimation (MINQUE) method with 1 for all the prior values, is suggested for estimating variance and covariance components in a bio-model for diallel crosses. Unbiasedness and efficiency of estimation were compared for MINQUE(l), restricted maximum likelihood (REML) and MINQUE(θ) which has parameter values for the prior values. MINQUE(l) is almost as efficient as MINQUE(θ) for unbiased estimation of genetic variance and covariance components. The bio-model is efficient and robust for estimating variance and covariance components for maternal and paternal effects as well as for nuclear effects. A procedure of adjusted unbiased prediction (AUP) is proposed for predicting random genetic effects in the bio-model. The jack-knife procedure is suggested for estimation of sampling variances of estimated variance and covariance components and of predicted genetic effects. Worked examples are given for estimation of variance and covariance components and for prediction of genetic merits.

We analyse patterns of the means and variances of genotypic fitnesses across different niches in a randomly mating haploid population. The population inhabits a spatially heterogeneous environment where it is subject to mutation and weak multilocus additive selection, with different selection coefficients in different niches. Approximate analytical expressions are derived for the stationary mean and variance of genotypic fitnesses among the niches in terms of environmental and genetic parameters. As a special case, we analyse an environment described by a variable t, distributed among the niches with mean t* and variance D*, and quadratic decrease in correlation between environments as a function of the difference in values of t. If the niches have the same qualities, the mean and variance of genotypic fitnesses evolve to be quadratic functions of t that achieve their maximum and minimum, respectively, at t*. With unequal niche qualities, these are non-polynomial functions that attain their extrema at different, usually intermediate values of t, although the coefficient of variation of the genotypic fitnesses still attains its minimum near t*. The functions involve the total mutation rate, the contribution of the loci to genotypic fitnesses, and the frequency and quality distributions of the niches. Thus, for this relatively simple model the norms of reaction may be calculated in terms of the detailed properties of the environmental heterogeneity, and the genetic system.

For a finite diploid population with no mutation, migration and selection, equations for the deviation of observed genotype frequencies from Hardy–Weinberg proportions are derived in this paper for monoecious species and for autosomal and sex-linked loci in dioecious species. It is shown that the genotype frequency deviation in finite random-mating populations results from the difference between the gene frequencies of male and female gametes, which is determined by two independent causes: the gene frequency difference between male and female parents and the sampling error due to the finite number of offspring. Previous studies have considered only one of the causes and the equations derived by previous authors are applicable only in the special case of random selection. The general equations derived here for both causes incorporate the variances and covariances of family size and thus they reduce to previous equations for random selection. Stochastic simulations are run to check the predictions from different formulae. Non-random mating and variation in census size are considered and the applications of the derived formulae are exemplified.

The exact test for neutrality based on the Ewens sampling distribution described previously (Slatkin, 1994) is not correct. The problem is that the test as described is based on the probability of the ordered configuration of numbers of alleles, while it should be based on the probability of the unordered configuration. The correctly implemented exact test leads to results that are similar to those from the homozygosity test proposed by Watterson (1977) for relatively small sample sizes but can still differ substantially for larger sample sizes. Programs to perform the exact test are available from the author.