The I factor is a LINE-like transposable element responsible for the I-R system of hybrid dysgenesis in Drosophila melanogaster. Inducer strains of this species contain several I factors whereas reactive strains do not. I factors are stable in inducer strains, but transpose at high frequency in the germ-line of females, known as SF females, produced by crossing reactive females and inducer males. Various abnormalities occur in SF females, most of which result from this high rate of transposition. We report here that recombination is increased in the germ-line of these females. This is a new characteristic of the I-R system of hybrid dysgenesis that might also be associated with transposition of the I factor.

In a random sample of 2263 cloned genomic DNA fragments of the wild potato species Solanum spegazzinii six related, highly repetitive fragments (SPG repeat family) were identified that were present in much higher copy numbers in S. spegazzinii when compared with the closely related cultivated potato S. tuberosum. The SPG repeat family was organized in long arrays of multiple copies. Cross hybridization experiments with 29 wild and cultivated Solanum species and with the related tomato showed specificity of the SPG repeat family for tuber-bearing Solanum species. Among tuber bearing Solanum species a high degree of variation was observed for restriction fragment length and copy number. The variation in copy number was not correlated with established taxonomic relationships between tuber-bearing Solanum species. DNA sequence analysis revealed a subrepeat structure of 120–140 base pairs embedded in longer repeat units of variable length. Length polymorphisms between highly repeated restriction fragments detected by the SPG probes were used for segregation- and linkage analysis in four mapping populations of potato, for which RFLP maps had been constructed. Twelve loci were identified, eleven of which mapped to the distal ends of nine linkage groups. All the evidence suggested that the SPG repeat family represents a satellite repeat members of which are localized in the subtelomeric region of potato chromosomes. The SPG repeat family could be used, therefore, for completing the genetic map of potato.

Ten recessive Opaque-2 (O2) alleles of independent origin were characterized at the molecular level. The results revealed a high level of polymorphism at the O2 locus. In addition, our data suggest the possible cause for the recessive character of some of the alleles investigated, and allow us to infer some conclusions concerning the degree of relationship between the o2 mutations. Comparison of genomic sequences spanning the first exon and obtained from a series of wild-type and recessive alleles revealed the presence of a hypervariable region, involving different dipeptides, in the N-terminal part of the O2 protein.

Seven imprinted genes are currently known in the mouse but none have been identified yet in the distal imprinting region of mouse Chromosome (Chr) 2, a region which shows striking linkage conservation with human chromosome 20q13. Both maternal duplication/paternal deficiency and its reciprocal for distal Chr 2 lead to mice with abnormal body shapes and behavioural abnormalities. We have tested a number of candidate genes, that are either likely or known to lie within the distal imprinting region, for monoallelic expression. These included 3 genes (Cebpb, E2f1 and Tcf4) that express transcription factors, 2 genes (Cyp24 and Pck1) that are involved in growth, 5 genes (Acra4, Edn3, Kcnb1, Mc3r and Ntsr) where a defect could lead to neurological and probably behavioural problems, and 3 genes (Cd40, Plcg1 and Rcad) that are less obvious candidates but sequence information was available for designing primers to test their expression. On/off expression of each gene was tested by reverse transcription–polymerase chain reaction (RT–PCR) analysis of RNA extracted from tissues of mice with maternal duplication/paternal deficiency and its reciprocal for the distal region of Chr 2. None of the 13 genes is monoallelically expressed in the appropriate tissues before and shortly after birth which suggests that these genes are not imprinted later in development. This study has narrowed down the search for imprinted genes, and valuable information on which genes have been tested for on/off expression is provided. Since there is considerable evidence of conservation of imprinting between mouse and human, we would predict that the 13 genes are not imprinted in human. Five of the genes: E2f1, Tcf4, Kcnb1, Cd40 and Rcad, have not yet been mapped in human. However, because of the striking linkage conservation observed between mouse Chr 2 and human chromosome 20, we would expect these genes to map on human chromosome 20q13.

Understanding how genetic variability is maintained in natural populations is of both theoretical and practical interest. In particular, the subdivision of populations into demes linked by low levels of migration has been suggested to play an important role. But the maintenance of genetic variation in populations is also often linked to the maintenance of sexual reproduction: any force that acts to maintain sex should also act to maintain variation. One theory for the maintenance of sex, the Red Queen, states that sex and variation are maintained by antagonistic coevolutionary interactions – especially those between hosts and their harmful parasites – that give rise to negative frequency-dependent selection. In this paper I present a model to examine the relationships between population subdivision, negative frequency-dependent selection due to parasites, the maintenance of sex, and the preservation of alleles from fixation. The results show strong interactions between migration rates, negative frequency-dependent selection, and the maintenance of variability for sexual and asexual populations.

Previous reports involving mammalian systems, particularly mice, have demonstrated the existence of cis- and trans-acting modifiers of transgene methylation. These modifiers are thought to be important in dominance modification, genome imprinting and cellular expression mosaicism. Their potential role in the penetrance and severity of many complex human diseases could be of even greater significance. In the present investigation we demonstrate that modifiers that act in a similar fashion to those identified in mice also exist in a non-mammalian vertebrate, the zebrafish Danio rerio. We also provide evidence that the transgene methylation pattern may be influenced by the sex of the individual and environmental modulators such as temperature and sodium butyrate. These data support the theory that this type of dominance modification is mechanistically similar to position effect variegation in Drosophila. Furthermore, these data suggest evolutionary conservation of the modifiers, at least within vertebrates, and imply that they and their actions are important in normal vertebrate development.

A loss of neutral genetic polymorphism is theoretically expected for many reasons in inbreeding organisms when compared to outbreeders. The first reason derives from the decrease of the effective population size, down to a halving, in purely selfing species. Other genetical reasons include hitchhiking and background selection. A loss can also be caused by ecological processes, that is by any kind of process provoking a genetic bottleneck. These theoretical expectations have been empirically confirmed in hermaphroditic plants for which selfing species exhibit both a lower gene diversity and number of alleles per population. Here I extend the analysis to hermaphroditic animals, mainly terrestrial and freshwater snails. The decrease of variability in selfers is far greater than what is expected under the halving of the effective size of populations only. Hitchhiking and background selection certainly cannot be rejected as causes of this extra loss. Bottlenecks can clearly be invoked in tropical freshwater snails. However a crude theoretical analysis using Ewens's sampling formulae shows that the relative loss of variability estimated by the number of alleles is smaller in inbreeders than in outbreeders assuming populations with the same number of individuals. This suggests that bottlenecks contribute less to the loss in selfers than in outcrossers. Variability lost within selfing populations of hermaphroditic animals is also lost at the level of a group of populations (metapopulation). This is theoretically not always expected. Indeed, I calculate the ratio of the effective size of a selfing metapopulation to be greater than that of an outcrossing one using previously derived equations. The large variation of this ratio which depends on both migration and effective size of subpopulations prevents prediction of the relative amount of genetic variability stored by selfing and outcrossing metapopulations.

Restriction enzyme molecular variation in Drosophila melanogaster Adh was compared between three natural populations from Europe, West Africa and East Africa. The frequency distribution of silent variation in the slow allele was compatible with the neutral model in all three samples. The number of haplotypes in East Africa was significantly higher than in the other two populations. The largest divergence, as measured by Fst, was between the East African population and a group made up from the West African, the European, and previously studied American populations. We suggest that a split first occurred within African populations at least 44000 years ago. European populations separated from West Africa more recently, between the last glacial maximum and the post-glacial optimum, 18000 to 8000 years ago. We suggest that this species was domesticated recently relative to human evolution, possibly with the advent of agriculture. Population differentiation with respect to the two allozymes, fast and slow, does not follow the geographical pattern of silent variation. It opposes European to both African populations, and probably results from selection for adaptation to alcohol in recent temperate populations.

We have determined the map position of the gene encoding a common precursor protein for diapause hormone and pheromone biosynthesis-activating neuropeptide (the DH-PBAN gene, Dh)in the silkmoth, Bombyx mori. First we compared the structure of introns in the DH-PBAN gene by the polymerase chain reaction, and found that the Dh locus carried three alleles, DhA1, DhA2 and DhB. The DhA1 and DhA2 alleles contained a fourth intron consisting of 740 bp, whereas DhB had a longer fourth intron of 770 bp. DhA1 and DhA2 contained a fifth intron consisting of 940 bp, whereas the fifth intron in DhB was much longer and consisted of 1700 bp. DhA1 was distinguished from DhA2 by an RFLP in the fifth intron after digestion with Rsa I. Linkage analyses using these polymorphisms showed that Dh was linked to the bp gene on chromosome 11, and independent of markers on chromosomes 1, 2, 3, 4, 5, 6, 7 and 13. To determine the map position, we obtained F1 hybrids between the n501 strain (K DhA1) and the w30 strain ( + KDhB), and backcrossed the F1 hybrid to females of the w30 strain. From the segregation of K and Dh in 864 individuals in the next generation, the recombination value was calculated as 25·5 % between K and Dh. Similarly we obtained backcross progeny between the No. 744 strain (BuDhA1) and the w30 strain ( + BuDhB), and calculated the recombination value between Bu and Dh as 30·4% from 487 progeny. Because k and Bu had already been mapped at positions 11–23·2 cM and 11–28·7 cM, respectively, we mapped Dh at 11--2·2 cM. The Dh locus is different from any loci which are known to control diapause, development or growth.

Previous theoretical studies of the evolution of the selfing rate have shown that mixed mating systems are not evolutionary stable states. Such models have, however, not included the effects of population structure and thus biparental inbreeding together with the evolution of selfing rates and inbreeding depression. In order to examine selection on selfing rates in structured populations, a stochastic model simulating a finite population with partial selfing and restricted pollen and seed dispersal has been developed. Selection on the mating system was followed by introducing modifiers affecting the selfing rate. The major result was that, with density dependent recruitment, a process which maintains the population structure necessary for biparental inbreeding to occur, a mixed mating system could be maintained. This result was associated with an increase of the mutation load with high selfing rates, and the selected selfing rate depended on the degree of population structure rather than on the initial selfing rate. With low dominance of deleterious alleles, complete allogamy can be selected for. Further studies showed that the more general condition of spatial heterogeneity of recruitment can lead to similar results, the most important condition being the maintenance of genetic structure within populations. A brief survey of the empirical literature shows that a positive relationship between the magnitude of inbreeding depression and the inbreeding coefficient within populations has been observed, in support of the present model.

Two series of 12½ day mouse chimaeric conceptuses were produced by aggregating (C57BL × CBA)F2 strain preimplantation embryos with embryos that differed at the Gpi-1s locus that encodes glucose phosphate isomerase, GPI-1. The composition of individual issues was evaluated by quantitative electrophoresis to estimate the % GPI-1A in the chimaeric tissue containing GPI-1A and GPI-1B. In one series of chimaeras, the GPI-1A cells were derived from a backcross between inbred BALB/c strain females and (BC × BALB/c)F1 males, where BC is the partly congenic strain C57BL/Ola.AKR-Gpi-lsa,c/Ws. In the other series of chimaeras, the GPI-1A cells were derived from the reciprocal backcross between (BC × BALB/c)F1 females and inbred BALB/c strain males. The [(BC × BALB/c)F1 female × BALB/c male] ↔ (C57BL × CBA)F2 series of chimaeras was reasonably balanced so that GPI-1 A and GPI-1B cells were fairly equally represented in the foetuses, placentas and extraembryonic membranes (tissue means: 37–51 % GPI-1A). This series did not differ significantly in composition from an earlier series of (BC × BALB/c)F2 ↔ (C57BL × CBA)F2 chimaeras. However, the [BALB/c female × (BC × BALB/c)F1 male] ↔ (C57BL × CBA)F2 series of chimaeras was unbalanced, with mean tissue compositions (28–33% GPI-1A) that were intermediate between the above two balanced series and the unbalanced (BALB/c × BALB/c) ↔ (C57BL × CBA)F2 series (tissue means: 14–22% GPI-1 A), that was studied previously. Thus, both (BALB/c×BALB/c) and [BALB/c×(BC x BALB/c)F1 embryos contributed less to the tissues of chimaeric conceptuses than either (BC × BALB/c)F2or [(BC × BALB/c)F1 × BALB/c] embryos. This implies that embryos from BALB/c mothers contributed less to the tissues of chimaeric conceptuses than embryos from (BC × BALB/c)F1 mothers. We, therefore, conclude that a maternal genetic effect is responsible for some of the differences in composition among the four groups of chimaeras. This maternal effect must act before the 8-cell stage but it is not yet known whether it is mediated via cytoplasmic inheritance, genomic imprinting or by the reproductive tract. Evidence that a maternal effect retards preimplantation development of embryos from BALB/c females is reviewed and the possibility that this might cause them to contribute poorly to chimaeric conceptuses when aggregated with more precociously developing embryos is discussed.

X chromosome inactivation (XCI) has been assumed to be complete in all cells of female mouse embryos at about 6 d post coitum (dpc). However, a recent study on β-galactosidase expression of an X-linked lacZ transgene suggests that XCI is probably not complete several days after this time in some lineages. To help resolve this issue, we analysed XCI in embryos which carry the T(X;16)16H (Searle's) translocation and are heterozygous at the X-linked Hprt and Pgk-1 genes. The quantitative RT-PCR single nucleotide primer extension (SNuPE) assay was used to measure Hprt and Pgk-1 allele-specific transcripts in embryos 9·5 dpc. No transcripts from the normal X chromosome were found in any of the tissues tested, indicating that inactivation was complete for these endogenous genes.

This paper describes the results of assays of male life-history characters in a large outbred laboratory population of D. melanogaster. Lines of flies homozygous for the entire third chromosome and lines of flies carrying two different third chromosomes were assayed for agespecific male mating ability (MMA), age-specific survivorship, male fertility, and body mass. The results of these assays were used to calculate the inbreeding decline associated with each of these traits, the average dominance of deleterious alleles that affect the traits, the genotypic and environmental components of variance for the homozygous lines, and phenotypic and genotypic correlations among the characters. Significant inbreeding decline was found for all characters except the Gompertz intercept and fertility. Early and late MMA show larger effects of inbreeding than any other trait. The inbreeding load for MMA is about the same magnitude as that for egg-to-adult viability, but is substantially less than that associated with total fitness. The estimated inbreeding decline and average dominance of male life-history characters are comparable to estimates for other Drosophila fitness components.

This paper studies the influence of two opposite forces, unidirectional unconditionally deleterious mutations and directional selection against them, on an amphimictic population. Mutant alleles are assumed to be equally deleterious and rare, so that homozygous mutations can be ignored. Thus, a genotype is completely described by its value with respect to a quantitative trait x, the number of mutations it carries, while a population is described by its distribution p(x) with mean M[p] and variance V[p] = σ2[p]. When mutations are only slightly deleterious, so that M » 1, before selection p(x) is close to Gaussian with any mode of selection. I assume that selection is soft in the sense that the fitness of a genotype depends on the difference between its value of x and M, in units of σ. This leads to a simple system of equations connecting the values of M and V in successive generations. This system has a unique and stable equilibrium, where U is the genomic deleterious mutation rate, δ is the selection differential for x in units of σ, and p is the ratio of variances of p(x) after and before selection. Both δ and ρ are parameters of the mode of soft selection, and do not depend on M or V. In an equilibrium population, the selection coefficient against a mutant allele is ŝ = δ2[U(2–ρ)]−1. The mutation load can be tolerable only if the genome degradation rate υ = U/σ is below 2. Other features of mutation-selection equilibrium are also discussed.

Methods for estimating probability of identity by descent (f) are derived for data on numbers of genotypes at single loci and at pairs of loci with many alleles at each locus. The methods are general, but are specifically applied to data on genotype frequencies in zygotes of the malaria parasite sampled from its mosquito host in order to find the extent of outcrossing in the parasite and the degree of clonality in populations. It is assumed that zygotes are the outcome either of gametes of the same clone, in which they are identical at all loci, or are products of two random, unrelated clones. From the estimate of f an effective number of clones per human host can also be derived. For Plasmodium falciparum from a Tanzanian village, estimates of f are 0·33 from data on zygote frequencies at two multiallelic loci, indicating that two-thirds of zygotes produce recombinant types.

A general representation of multilocus selection is extended to allow recombination to depend on genotype. The equations simplify if modifier alleles have small effects on recombination. The evolution of such modifiers only depends on how they alter recombination between the selected loci, and does not involve dominance in modifier effects. The net selection on modifiers can be found explicitly if epistasis is weak relative to recombination. This analysis shows that recombination can be favoured in two ways: because it impedes the response to epistasis which fluctuates in sign, or because it facilitates the response to directional selection. The first mechanism is implausible, because epistasis must change sign over periods of a few generations: faster or slower fluctuations favour reduced recombination. The second mechanism requires weak negative epistasis between favourable alleles, which may either be increasing, or held in check by mutation. The selection (si) on recombination modifiers depends on the reduction in additive variance of log (fitness) due to linkage disequilibria (υ1 < 0), and on non-additive variance in log (fitness) (V′2, V′3,.. epistasis between 2, 3.. loci). For unlinked loci and pairwise epistasis, si = − (υ1 + 4V2/3)δr, where δr is the average increase in recombination caused by the modifier. The approximations are checked against exact calculations for three loci, and against Charlesworth's analyses of mutation/selection balance (1990), and directional selection (1993). The analysis demonstrates a general relation between selection on recombination and observable components of fitness variation, which is open to experimental test.

Cumulus-oocyte complexes from the two consomic strains of mice, B10. BR and BIO.BR-Ydel (with a partial deletion of the Y chromosome) and the BALB/c strain, as well as reciprocal Fl and F2 hybrids, were tested for their susceptibility to attack by hyaluronidase. The cumulus cell dispersal was significantly more rapid in the eggs of B10.BR females and of Fl and F2 hybrid females sired by the B10.BR males than in those of BALB/c females and hybrids sired by BALB/c males. These results confirm the earlier data of Bander et al. (1989) which were interpreted as evidence of paternal imprinting of the C57BL genome. In contrast, cumulus cell dispersal in the eggs of females sired by BIO.BR-Ydel males was significantly slower than in females sired by B10.BR males, and did not differ from that in the BALB/c strain. The results suggest that the partial deletion of the Y chromosome abolished the effect of paternal genome imprinting which is observed in the B10.BR strain.

We study the transient dynamics of the genotypic variance of an additive trait under stabilizing selection, recombination and random drift. We show how interaction of these factors determines the form and the rates of change of different components of the genotypic variance. Let Vg be the genie variance of the trait and CL be the contribution of linkage disequilibrium to the genotypic variance. We demonstrate that the dynamics of the system on the plane (Vg, CL) are typically characterized by a quick approach to a straight line with slow evolution along this line afterwards. We show that the number of loci, n, and the population size, N, affect the expected dynamics of Vg mainly through the ratio N/n. We use our analytical and numerical results in interpreting the published results of artificial stabilizing selection experiments. The analysis suggests that it is drift and not selection that most likely led to the reduction of genetic variability in most of these experiments. Even very strong stabilizing selection only slowly removes polygenic variability from populations.