The genotypes are given of over 4500 asci showing aberrant segregation at the grey (g) spore colour locus in Sordaria fimicola from crosses, with outside markers present, between g locus mutants and wild type, or involving pairs of g locus mutants that differed in spore colour (g locus mutants range from grey to hyaline). This colour difference enabled asci with aberrant segregation of the mutant spore colours to be detected in the two-point crosses, as well as those with wild-type (black) spores. From a total of 21 one-point crosses involving altogether seven primary and two secondary g locus mutants, 2700 aberrant asci have been grouped into the 14 ascus genotypes that can be recognized, and in the two-point crosses 1830 asci (1044 fully scored) from 10 pairwise combinations of the mutants have been classified into the 150 ascus genotypes that can be distinguished.

Interallelie crosses of mutants at the grey (g) spore colour locus in Sordaria fimicola, heterozygous for flanking markers, give rise to a large number of aberrant ascus genotypes, 45 of which can arise through relatively simple events and have been chosen for study. These genotypes comprise 50–75% of the aberrant asci, depending on the mutants crossed.

Comparison of the results from 10 pairwise crosses involving 7 alleles reveals that linked postmeiotic segregation and co-conversion decrease rapidly in frequency with increasing separation of the mutant sites.

The data from reciprocally recombinant asci, from asci with normal segregation at one of the two mutant sites, and from flanking marker behaviour in one- and two-point crosses, agree with the Holliday-Sobell formulation, with the following additional features:

(1) The nuclease, which nicks homologous polynucleotides and then degrades one of the two nicked chains when a mutant enters the hybrid DNA structure, can show preferential degradation of the mutant (or the wild-type) chain. In addition, a second nuclease is involved in the excision-repair process that introduces an additional preferential (marker specific) bias in the degradation of the mutant (or the wild-type) chain. This could explain why asci with odd-ratio conversion (5:3 and 3:5 ratios) sometimes show a different bias, as first reported by Emerson for Ascobolus, from those with even-ratio conversion (6:2 and 2:6 ratios), since the latter but not the former require, in addition, the action of a mismatch correction enzyme to account for them.

(2) The migratory hybrid DNA structure which enters the gene at one end may be of a different size from that which enters from the other end.

(3) Mismatch correction at the end of the hybrid DNA structure leads to a non-recombinant outside marker genotype and modifies the 1:1 ratio of parental:recombinant flanking markers that is otherwise found.

For large population sizes, gene frequencies p and q at two linked over-dominant loci and the linkage disequilibrium parameter D will remain close to their equilibrium values. We can treat selection and recombination as approximately linear forces on p, q and D, and we can treat genetic drift as a multivariate normal perturbation with constant variance-covariance matrix. For the additive-multiplicative family of two-locus models, p, q and D are shown to be (approximately) uncorrelated. Expressions for their variances are obtained. When selection coefficients are small the variances of p and q are those previously given by Robertson for a single locus. For small recombination fractions the variance of D is that obtained for neutral loci by Ohta & Kimura. For larger recombination fractions the result differs from theirs, so that for unlinked loci r2 ≃ 2/(3N) instead of 1/(2N). For the Lewontin-Kojima and Bodmer symmetric viability models, and for a model symmetric at only one of the loci, a more exact argument is possible. In the asymptotic conditional distribution in these cases, various of p, q and D are uncorrelated, depending on the type of symmetiy in the model.

Selection for high and low growth-rate was carried out during eight generations of asexual propagation by zoospores and seven generations of sexual reproduction by oospores. The fungus has previously been shown to be diploid during its vegetative phase. In the zoospore lines there was no significant variation and no response to selection, except for the occasional appearance of fast-growing sectors. A high line was established from such a sector; in its sexual progeny the inheritance of growth-rate was non-Mendelian. Propagation through self-fertilized oospores released very considerable genetic variation, and both high and low lines responded to selection. At first the variation within families, and the response to selection, increased with succeeding generations, despite the intense inbreeding. In later generations the high line became less variable, and the progeny oospore cultures resembled the fast-growing sectors. It is concluded that growth-rate is controlled by a polygenic system and by cytoplasmic determinants, a mutant form of which is responsible for the fast-sectoring phenotype.

Clones of V79 Chinese hamster cells resistant to the steroid compound ouabain have been induced by mutagen treatment with MNNG (= N-methyl-N′-nitro-N-nitrosoguanidine). The frequency of resistant colonies was increased from spontaneous value of 2 × 10−6 to a frequency of 1·2 × 10−3 with MNNG treatment.

The highest number of induced mutants appeared when an expression time pf 40 h was allowed between MNNG treatment and addition of ouabain.

A new metric character in mice, ‘footpad bristle number’ is described. The character is easily measured under a stereo-microscope, is very highly inherited, and is independent of age, sex and the environment common to littermates. Such a character might prove useful in a number of quantitative genetic studies in mice.

Klebsiella aerogenes V9A contains two lac operons, one chromosomal and one borne by the Lac plasmid, FKlac. It is shown that the plasmid Lac repressor resembles that of Esherichia coli in that it represses the E. coli operon but is inactivated by melibiose. In contrast, the chromosomal Lac repressor of V9A is unable to repress the E. coli operon or the FKlac operon, but is converted into an active repressor of both these operons on becoming bound to melibiose. The chromosomal Lac repressor is able to repress its own operon whether or not it is bound to melibiose, but is inactivated by isopropyl-β-D-thiogalactoside. These results support the hypothesis that the Klebsiella Lac plasmid was derived from an ancestral E. coli chromosome.

The tetracycline-resistance determinants of R-factors from different compatibility groups have been tested in Escherichia coli K12 and phenotypically classified into two major classes. Class I determinants confer high-level resistance to tetracycline (> 100 μg/ml) and moderate resistance to minocycline (5–25 μg/ml) while those of Class II gave moderate resistance to tetracycline (50–70 μg/ml) and low resistance to minocycline. Each class was subdivided because of variation in resistance profiles and in the abilities of tetracycline and minocycline to induce increased resistance. Strains carrying two compatible Tetr R-factors of the same or different phenotypic groups did not show increased tetracycline resistance.