The stability of conversion frequencies with time was investigated in Ascobolus immersus. There were usually marked reductions in gene conversion frequencies at locus wl as crosses matured, to about one third of the initial values. This applied to all six wl alleles tested, irrespective of their molecular nature, and at all temperatures used. Asci dehiscing early from an apothecium had much higher conversion frequencies than those dehiscing late, but there were no differences between apothecia maturing at different times within a cross. Alleles at four loci unlinked to wl were also tested. All four showed significant changes, though not in all crosses; three loci showed decreases in conversion frequency with time, while one showed an increase. The relative frequencies of different conversion classes often changed with time. These changes appear to result from alterations in locus-specific recombination initiation and in repair of base mispairs in hybrid DNA, not from differential maturation rates of different ascal segregation classes. These effects could cause misinterpretation of quantitative tests of recombination models from gene conversion data.

The B mutation is associated with a tandem duplication of 16A1–16A7. It is unstable, mutating to wild type and to a more extreme form at a frequency of one in 1000 to 3000. The reversion to wild type is associated with the loss of one copy of the duplication, whereas the mutation to extreme B is associated with a triplication of the region. The instability of B has been attributed to unequal crossing-over between the two copies of the duplication. Recent molecular data show that there is a transposable element, B104, between the two copies of the duplication and support the hypothesis that this element generated the duplication via a recombination event. These data suggest that unequal crossing-over within the duplication may not be the cause of the instability of B. Instead, the instability may be caused by a recombination event involving the B104 element. This issue was addressed using probes for the DNA on either side of the B104 element at the B breakpoint. All of the data indicate that the B104 element is not involved in the instability of B and support the original unequal crossing-over model.

We conducted classical genetic analysis of the difference in male genitalia and hybrid sterility between the island-dwelling sibling species Drosophila sechellia and D. mauritiana. At least two loci (one on each autosome) are responsible for the genital difference, with the X chromosome having no significant effect. In contrast, male hybrid sterility is caused by at least four gene loci distributed among all major chromosomes, with those on the X chromosome having the largest effect.

We also show that the large difference in ovariole number between D. sechellia and its mainland relative D. simulans is due to at least two gene substitutions, one on each major autosome. The X and the left arm of the second chromosome, however, have no significant effect on the character. This implies that the evolution of reduced ovariole number involved relatively few gene substitutions.

These results extend previous findings that morphological differences between Drosophila species are caused by genes distributed among all chromosomes, while hybrid sterility and inviability are due primarily to X-linked genes. Because strong X-effects on male sterility have been found in all three pairwise hybridizations among D. simulans, D. sechellia and D. mauritiana, these effects must have evolved at least twice independently.

Mitochondrial DNA (mtDNA) variation in Drosophila simulans was studied to determine whether the cytoplasmic state of mtDNA heteroplasmy persists in natural populations in Réunion. For this purpose, 172 isofemale lines, newly collected from two local populations, were examined, among which three types of mtDNA (siII, siIII and siIII′) were found, based on the Hpa II restriction pattern. Ten of the lines were heteroplasmic for a combination of siII and siIII, as determined by autoradiography. The same type of heteroplasmy had been noted in one of the two local populations 8 years before (Satta et al. 1988). The present results suggest that the heteroplasmic state occurs recurrently in natural populations of D. simulans in Réunion.

This paper examines models of the population dynamics of transposable elements when chromosomal sites vary with respect to the effect on fitness of mutations caused by element insertions. Element abundance is assumed to be stabilised solely by the joint results of transposition, excision, and selection against insertional mutations. When there are only two classes of site, selected and neutral, it is hard to find parameter values for which numbers of elements are maintained that match the findings from surveys of Drosophila populations, as elements tend to accumulate at high frequencies at the neutral sites. It is similarly hard to produce realistic equilibria with three classes of site (strongly selected, weakly selected, and neutral), when elements can transpose out of the neutral sites. If transposition from neutral sites is impossible, as might be the case for elements inserted into centric heterochromatin, then realistic equilibria can be generated if there is very weak selection against elements inserted into the majority of non-neutral sites. This model predicts a modest over-representation of elements at the neutral sites. It also predicts that elements should be under-represented on the X chromosome compared with the autosomes, but this is not generally found to be the case. It is concluded that selection against insertional mutations is unlikely to be the major factor involved in the containment of element abundance.

Segregation of sperm abnormality level and the pattern of major urinary proteins (MUPs) were investigated in F2 and B1 hybrid males obtained from crosses involving two contrasting inbred strains of mice: CBA/Kw (Mup-1a1a, 3·3% abnormal sperm) and C57BL/Kw (Mup-1b1b, 21·9% abnormal sperm). In the progeny of both crosses mean levels of abnormal spermatozoa were significantly higher for males typed as Mup-1b1b than for heterozygous Mup-1a1b males. Moreover, all F2 hybrid males showing very high percentages of abnormal sperm were Mup-1b1bhomozygotes. Similarly, among B1 males with a high level of deformed spermatozoa, a statistically significant majority were Mup-1b1b genotypes. Our results suggest that at least two genes which influence sperm abnormality level are segregating in these crosses. Both appear to be recessive for high sperm abnormality level, and one shows weak linkage to Mup-1 on chromosome 4.

A mutation resulting in increased triosephosphate isomerase (TPI) activity in blood was recovered in offspring of procarbazine hydrochloride-treated male mice. Breeding experiments indicated a codominant mode of expression. Compared to the wild type, heterozygous and homozygous mutants have mean erythrocyte TPI activities of approximately 140 and 190%, respectively. Besides blood and erythrocytes the increased activity is expressed to a similar degree in spleen, and to a lesser degree in liver, lung, kidney, muscle and brain. Enhanced activity was absent in the heart. Heterozygous and homozygous mutants are viable, fully fertile and exhibit no significant differences in haematological or other physiological traits studied. Biochemical investigations of TPI in both mutant genotypes revealed neither physicochemical nor kinetic differences compared to the wild type. Moreover, immunoinactivation studies showed no difference in the amount of antiplasma required to inactivate a constant amount of TPI activity in all three genotypes, strongly suggesting that the differences in enzyme activity are attributable to differing amounts of enzyme protein expressed per cell. Mapping studies indicated that the mutation is closely linked to the Gapd locus and consequently is located either adjacent to or within the Tpi-1 structural locus. It is hypothesized that the mutation affected a regulatory element contiguous to the Tpi-1 structural locus which acts by increasing the amount of TPI expressed.

The mouse t complex on chromosome 17 is known to harbour many genes which have an important role in spermatogenesis. One of these, Tcp-1 has been cloned and shown to code for a protein probably essential for acrosome formation. During the isolation of a cDNA for Tcp-1 two other homologous sequences were recognized and described as Tcp-1x and Tcp-1y. In this paper we describe the isolation of a cDNA which has been shown by in situ hybridization to correspond to the Tcp-1x gene. Sequence analysis has confirmed that a 140 bp region of homology between Tcp-1 and Tcp-1x lies in the 3′ portion of both genes. Northern blotting has revealed that the Tcp-1x gene is expressed abundantly in liver where two transcripts are detectable and hybrid selection shows that the gene codes for a 37 kDa protein. A search of the DNA databases has failed to find any significant homology between Tcp-1x and any other sequences apart from Tcp-1.

The effects of different types of insemination (normal and delayed matings and in vitro fertilization) on the transmission ratio distortion (TRD) of three t haplotypes were determined. The tw73 haplotype which contains all of the loci known to affect TRD is transmitted at equivalent frequencies in normal matings and in in vitro fertilizations (0·84 and 0·85, respectively) but at a significantly lower frequency (0·62) in delayed matings. The distal partial th18 haplotype is transmitted at equivalent frequencies in all types of insemination (0·66 to 0·70) while the proximal partial tw18 haplotype is transmitted in Mendelian frequencies in normal matings and in in vitro inseminations but at a significantly lower frequency in delayed matings. The results are discussed with reference to the current genetic model for transmission ratio distortion.

Deterministic computer calculations were used to investigate the effects on the fitnesses of genotypes at neutral loci that are caused by associations with several linked or unlinked selected loci, in partially self fertilizing populations. Both mutation to partially recessive alleles and heterozygote advantage at the selected loci were studied. In the heterozygote advantage models, either arbitrary linkage between all loci was modelled, with a single neutral locus, or many unlinked selected and neutral loci were modelled. Large apparent overdominance could be generated in all types of model studied. As has previously been suggested, these types of effect can explain the observed associations between fitness and heterozygosity in partially inbreeding populations. There were also apparent fitness differences between the genotypes at the neutral locus among the progeny produced by selfing, especially with linkage between the neutral and selected loci. There is thus no genotype-independent fitness value for these progeny. Marker based methods for estimating the relative fitness of selfed and outcrossed progeny assume equality of these fitnesses, and will therefore be inaccurate (with in most cases a bias towards overestimating the degree of inbreeding depression) when there is linkage between the neutral marker loci and loci determining fitness.

Mean fitness and inbreeding depression values in multi-locus models of the control of fitness were studied, using both a model of mutation to deleterious alleles, and a model of heterozygote advantage. Synergistic fitness interactions between loci were assumed, to find out if this more biologically plausible model altered the conclusions we obtained previously using a model of multiplicative interactions. Systems of unlinked loci were assumed. We used deterministic computer calculations, and approximations based on normal or Poisson theory. These approximations gave good agreement with the exact results for some regions of the parameter space. In the mutational model, we found that the effect of synergism was to lower the number of mutant alleles per individual, and thus to increase the mean fitness, compared with the multiplicative case. Inbreeding depression, however, was increased. Similar effects on mean fitness and inbreeding depression were found for the case of heterozygote advantage. For that model, the results were qualitatively similar to those previously obtained assuming multiplicativity. With the mutational load model, however, the mean fitness sometimes decreased, and the inbreeding depression increased, at high selfing rates, after declining as the selfing rate increased from zero. We also studied the behaviour of modifier alleles that changed the selfing rate, introduced into equilibrium populations. In general, the results were similar to those with the multiplicative model, but in some cases an ESS selfing rate, with selfing slightly below one, existed. Finally, we derive an approximate expression for the inbreeding depression in completely selfing populations. This depends only on the mutation rate and the dominance coefficient and can therefore be used to obtain estimates of the mutation rate to mildly deleterious alleles for plant species.

Two congenic lines of mice, one with a partial deletion of the Y chromosome, differ in the percentage of spermatozoa with abnormal heads: B10.BR/SgSn males give 22·6%; and B10.BR-Ydel/Ms males give 64·2%; abnormal sperm. The F1s resulting from crosses of B10.BR/SgSn males with females of five common inbred strains exhibited significantly lower levels of abnormal sperm than the parental strains, as opposed to F1 hybrids sired by B10.BR-Ydel/Ms mutant males, where very high levels of abnormal spermatozoa were found. About 30%; of abnormal spermatozoa, produced by males with deletion on the Y chromosome, were characterized by a flat acrosomal cap. This class of abnormality was never observed in non-mutant males, suggesting a mutant-specific defect. These results demonstrate the important role of the Y chromosome in spermatogenesis.

Genet. Res., Camb. (1991), 57, pp. 51–54

page 53, 2nd column: the last word in line 4 should read superior, not inferior.