A sex-linked mutation with a phenotype similar to that of an allele of the Mottled series has been discovered in a mosaic male. This mutation can be normally transmitted through the sperm of the original male but not through the oocytes. Several interpretations of this abnormal transmission are discussed.

The variation in the response to selection was studied by replication of selected lines. A random-bred strain of mice was divided into six replicates. Two-way selection for 6-week weight was applied in an identical manner to each replicate, and each had an unselected control. Each line (6 large, 6 control, 6 small) was maintained by minimal inbreeding with 8 single-pair matings. The overall mean responses, both up and down, were linear and very regular for ten generations, with realized heritabilities of 40% upwards, 33% downwards and 37% for the divergence. The separate replicates, however, differed greatly in their realized heritabilities, with upward selection ranging from 25 to 46%, and downward selection from 16 to 50%. The theoretical prediction that, because of genetic drift, the standard error of a realized heritability is underestimated by the standard error of the regression of response on cumulated selection differential was borne out in this experiment. The empirical standard error, calculated from the observed variance between replicates, was more than twice as great as that of the regression. The empirical standard errors showed that the asymmetry between upward and downward responses was not significant. The variation between the replicates was ascribed mainly to random drift, which may seriously influence the conclusions about the realized heritability and the asymmetry of response that would be drawn from a single experiment with the population size of one of these replicates. After 23 generations of selection the large lines were approaching limits, and the limit appeared to be at the same level in all. The small lines showed an undiminished realized heritability after 23 generations, but the selection differentials were then so small that little progress was made. There was evidence of counter-acting natural selection. All aspects of productivity – proportion of fertile matings, litter size and weaning rate – declined in the control lines. The overall productivity of the large lines was a little below the controls, and that of the small lines was reduced to about half the level of the controls. The separate replicates differed from each other significantly in all the components of productivity.

The P–M system of hybrid dysgenesis in Drosophila melanogaster is a syndrome of genetic abnormalities which appears in the progeny of crosses between strains different with regard to their possession of ‘P’ transposable elements. Cytotype is an extrachromosomal property which regulates the mobility of the P element. We report here data showing that a cytotype polymorphism previously observed in a natural population from North-Africa is stable over a period of 5 years. A potentially high rate of mutation is associated with this cytotype polymorphism. Explanations of the appearance of a cytotype polymorphism are proposed and the consequences for the genetic load induced by transposable elements are discussed.

Non-linear offspring-parent regressions and heritabilities are expected for characters showing genetic asymmetry due to directional dominance and/or asymmetrical gene frequencies. Since reproductive fitness characters exhibit these characteristics, they should show consistently nonlinear heritabilities, with greater heritabilities in the direction of lower fitness. As a consequence, responses to bi-directional selection on fitness traits should be asymmetrical in the same direction. This prediction has been tested by an analysis of published bi-directional selection experiments for reproductive fitness traits. Significant asymmetry (24 of 30 studies) in the predicted direction was found. For studies reporting realized heritabilities, the means were 0·173 and 0·259 for lines selected for higher and lower reproductive fitness, respectively, the high lines being 33% less than the low lines. Asymmetry was evident for studies reporting realized heritabilities and for those with random mating controls of the same size as the selection lines. Consequently, it is argued that the asymmetry results from genetic asymmetries. This asymmetry has important implications in the improvement of reproductive fitness traits in plant and animal breeding.

A suite of aberrant genetic traits, including increased mutation rate, sex-limited mutation and distorted transmission ratios, was produced among progeny of genetic crosses between two strains of Drosophila mauritiana when a paternally contributed Mos excision factor is placed into a non- Mos genetic background. In the reciprocal cross, involving maternally contributed Mos and Mos associated cytoplasm, the same genetic abnormalities are not observed. Differential effects on mariner excision in germ-line versus somatic tissue are apparent. Because Mos is known to influence the mobility of the mariner transposable element, these traits may be associated with mariner excision and/or transposition.

The suppressor of sable [su(s)2] restores the function of vermilion (v), purple (pr) and speck (sp) as well as sable (s) in Drosophila melanogaster. In this report various alleles of su(s) are compared for their relative effectiveness on three target mutations, v, pr and sp. Three criteria for suppression of pr and v were employed: visible phenotype, eye pigment levels (drosopterins and xanthommatin) and enzyme levels (sepiapterin synthase and tryptophan oxygenase). For sp only the visible phenotype was examined. By all three criteria pr was found to be more easily suppressed than v; v and sp were comparable. By use of pr with various alleles of su(s) either homozygously or in heterozygous combination with su(s)+, the extent of suppression of pr can be best demonstrated by observing the levels of sepiapterin synthase; normal levels of drosopterins were found in females when sepiapterin synthase was only 20% of normal. On the other hand, the extent of suppression of v is best demonstrated by the amount of xanthommatin eye pigment, because even the suppressed vermilion fly has < 10% of wild-type activity of tryptophan oxygenase when 1-day-old flies are examined; in older flies this enzyme can be as high as 50% of wild type. From these results we also demonstrated that su(s)2, and other alleles, are not recessive but, in heterozygous combination with su(s)+, cause marked suppression of pr and slight, but reproducible, suppression of v. The purple mutation, therefore, is particularly useful for studying the mechanism of suppression as well as for obtaining new mutant alleles of su(s).

Current models based on the analysis of linear metabolic pathways at steady-state predict that large increases over wild type in the activity of one enzyme will not alter an organism's fitness. This prediction is tested at steps in a highly branched pathway under two conditions known to alter steady-state: heat shock and nitrogen starvation. Saccharomyces cerevisiae transformants overproducing 1 of 4 enzymes in glycolysis (hexokinase B, phosphoglucose isomerase, phosphofructokinase, or pyruvate kinase) were subjected to heat shock in both exponential and stationary phases of growth. In neither phase does enzyme overexpression alter heat shock sensitivity. When starved for nitrogen in acetate medium, transformants overproducing hexokinase, phosphoglucose isomerase, and phosphofructokinase sporulate at the same rate and with the same frequency as cells harbouring only the plasmid vector. Current models therefore correctly predict the relationship between activity and components of fitness for 3 of 4 enzymes. By contrast, cells overexpressing pyruvate kinase sporulate poorly. This defect is not observed among cells transformed with a plasmid containing a Tn5 disrupted copy of the PYK gene. These findings are consistent with reports that implicate the PYK locus in yeast cell cycle control and suggest that it may be challenging to model relations between fitness and activity for multifunctional proteins.

In these experiments comparisons were made between the magnitude of the interaction of ‘pure’ strains and strain crosses of poultry with two types of environments—location effects and a restricted-feed versus a full-feed rearing programme. The ‘pure’ strains were closed flocks of White Leghorns that had been selected for increased egg production, while the strain crosses were the reciprocal crosses of all combinations of these pure strains. Data from four separate experiments in four consecutive years used for this study involved 8320 laying birds. Six traits of the adult laying birds were used for these analyses.

It was expected that the ‘pure’ strains would differ in performance amongst themselves to a greater extent than the strain crosses, and for the two traits, body-weight at housing and sexual maturity, this was found to be the case in three out of four years. These two traits were affected to the greatest extent by the rearing treatment. Also, the genotype-environment interaction variance was found to be significant and of important magnitude relative to the genetic variance for these two traits. Where the environmental effect was found to be smaller, the interaction variance made up a smaller proportion of the genetic variance.

The tw5 haplotype is a recessive mutation which is lethal when homozygous in mouse embryos following implantation. This series of studies was undertaken to determine the effect of the tw5/tw5 genotype on embryos developing in vitro. Blastocyst embryos from + / tw5inter se matings were compared with control blastocysts obtained from matings between T/ + and + / + females and + / tw5 males for their abilities to continue development in vitro in two culture media. The data show that there are no significant differences between the percentages of experimental and control blastocyst embryos which attach and outgrow or which contain inner cell masses on any day of culture up to equivalent gestation day 21 in either media. These findings show that the life span of cells from tw5 / tw5 embryos can be extended significantly by in vitro culture.

The Drosophila melanogasterecd1 mutation causes a severe temperature-sensitive deficiency in the titre of the steroid hormone eedysone. This mutation was used to investigate the role of eedysone in both the transcription of the genes mapped at the 3C11–12 intermoult puff region and the puff formation. Thoroughly synchronized ecd1 larvae were shifted to the non-permissive temperature at various times of the development; after 24 or 48 h, the levels of the transcripts derived from Sgs-4, Pig-1 and ng-1, the three genes located at the 3C11–12 polytene bands, were determined. The results showed that the levels of the transcripts encoded by Pig-1 and ng-1 are unaffected by the drop in the ecdysone titre occurring in non-permissive conditions whereas the amount of Sgs-4 mRNA is greatly reduced. These data clearly indicate that transcription of the three genes mapped within the puff region is affected differently by the hormone. Furthermore, ecd1 larvae cultured at the non-permissive temperature show a prominent puff at the 3C11–12 polytene bands, indicating that eedysone is not essential for puff induction and that puff size is not simply correlated with high-level Sgs-4 transcription.

Lysogenization by phage P22 carrying amber mutations in genes 12 and 18

By Amiram Ronen and Myron Levine

Department of Genetics, The Hebrew University of Jerusalem, Israel, and the Department of Human Genetics,

The University of Michigan, Ann Arbor, Michigan 48109, U.S.A.

Vol. 27, No. 2, p. 337: add to Acknowledgements –

‘In addition, this research was supported by Grant No. 738 from the United States–Israel Binational Science Foundation (BSF), Jerusalem, Israel.’

Mutants of Aspergillus flavus impaired in aflatoxin production were induced with N-methyl-N′-nitrosoguanidine and analysed by means of the parasexual cycle. The gene symbol afl was assigned to this type of mutation. Diploid complementation tests revealed that most of 14 afl mutants belonged to different complementation groups. One mutant (afl-1) failed to complement or only partially complemented all other mutants. Haploidization of one diploid revealed the independent segregation of two afl mutants. Heterozygous diploids were synthesized between two afl mutants and tester strains genetically labelled on eight linkage groups. Haploidization of the diploids led to the assignment of two aft mutants to linkage groups. Linkage of afl-4 to w+ and afl-1 to leu on linkage groups II and VII, respectively, was demonstrated.