In a two-dimensional stepping-stone model of finite size, if a pair of alleles happen to segregate in the whole population, marked local differentiation of gene frequencies can occur only if migration between colonies is sufficiently rare so that Nm < 1, where N is the effective size of each colony and m is the rate at which each colony exchanges individuals with four surrounding colonies each generation. On the other hand, if Nm ≥ 4, the whole population behaves as if it were panmictic and the allelic frequencies become uniform over the entire distribution range unless mutation is unusually high. Tendency toward local differentiation is much weaker in two-dimensional than in one-dimensional habitats.

The process leading to gene recombination can be interrupted in the filamentous bacteria Streptomyces coelicolor by growing mixed cultures on cellophane disks lying on complete medium. The mycelium is harvested, broken, diluted and the broken hyphae plated at different time intervals. By this means some markers can be excluded from heteroclones or from recombinant progeny in early samples. The recombinant pattern clearly changes with time, with an increase of markers contributed to the recombinant progeny. In crosses between male (NF) and female (UF) strains, the maleness is the first donor trait to appear in the cells of the recipient parent. The fertility factor does not produce a transfer origin on the donor chromosomes; the donor contribution may extend on either side or on both sides of the factor which appears to be compulsory for zygote formation. The longer the time of contact between parental cells, the longer the segment of the donor chromosome contributing to the recombinant progeny. When spores are formed they contain almost exclusively recombinant nuclei derived from segregation processes.

The rates of mitotic recombination at the loci ade3, ade5–7, ade6 and ade8 were approximated by the method of ‘increasing proportion of variants with growth’. The values (per cell generation) were 3·5 × 10−4 for ade8, 1·4 × 10−4 for ade5–7, 4·0 × 10−5 for ade3 and < 2 × 10−6 for ade6. The relative frequencies of mitotic recombination in regions between ade 5–7 and the centromere were different from those obtained with radiation-induced recombinants.

1. Coprinus lagopus produces two non-specific phosphatases: a constitutive acid phosphatase, and an alkaline phosphatase which is repressed during growth on media with a high inorganic phosphate concentration.

2. The alkaline phosphatase is also repressed when Coprinus is grown on an organic phosphate source; but if the acid phosphatase is selectively inhibited by fluoride the alkaline phosphatase is de-repressed and growth is comparable to that observed on an inorganic phosphate source.

3. Alkaline phosphatase is not repressed in aerial mycelium or sporophores even when grown on high phosphate medium.

4. Mutants altered in their capacity to synthesize alkaline phosphatase were selected from two compatible wild-type strains, H2 and H5.

5. Mutants producing a higher level of alkaline phosphatase than wild-type (‘regulator’ mutants) fall into four (or possibly five) complementation groups. Assuming five separate genes, two pairs are linked; the remaining one is independent and on another chromosome.

6. Mutants deficient in alkaline phosphatase synthesis fall into at least three groups. They were tested for linkage to ‘regulator’ loci but so far there is no evidence of this.

Hyperhaploid and hyperdiploid colonies of Aspergillus nidulans, disomic or trisomic for one or more chromosomes respectively, are abnormal in phenotype and somatically unstable. It has been shown for Glasgow strains that a different specific phenotype arises, depending upon which chromosome is additional (Pollard, Käfer & Johnston, 1968). This report extends this system and shows that irrespective of the residual genetic background of the isolate, the system of phenotypic specificity applies.

R-factor 1818, which is refractory to curing by acridines or ethidium bromide from Escherichia coli K 12 and Klebsiella aerogenes 418 hosts, was eliminated from thymineless mutants of these organisms under conditions of thymine starvation. Neither R-7268 nor R-TEM could be eliminated by this method. The concurrence of R-TEM with R-1818 inhibited the curing of R-1818, whereas R-7268 had no effect on R-1818 elimination

Genetic analysis of spontaneous half-sectored colonies of a diploid strain homozygous for ade2 and heterozygous for linked markers aro1A, hom2, ade8 and trp4 was undertaken. Some of these were due to an unexpectedly high rate of ade2 reversion and others to a low incidence of haploid sectors. About 50% resulted from mitotic recombination. Only a few were due to mitotic crossing over alone, the majority resulted from non-reciprocal ‘conversion’, including recombination extending over two loosely-linked loci, and some colonies featured both mitotic crossing over and mitotic conversion. The latter indicate that negative interference operates for mitotic recombination. The rate of mitotic recombination for ade8, calculated from the frequency of half-sectored colonies was 3·1 × 10−4 per cell division.

Mutants of Pseudomonas aeruginosa isolated as being resistant to p-fluorophenylalanine (FPA-r) are pleiotropic, showing patterns of phenotypic alteration of host-controlled modification, growth rate at 37 and 43 °C, ability to grow on minimal medium, response to antibiotics and ribosomal characteristics. Alterations in host-controlled modification are also found in FPA-r mutants of Escherichia coli B.

The pleiotropy arising in these FPA-r mutants of P. aeruginosa appears to result from alterations in the specificity of protein synthesis. Phenotypic variations in host-controlled modification of the type found in the FPA-r mutants may provide a method for the detection of suppressor mutations which act by miscoding.

Multiple recombinational events within a gene were studied by tetrad analysis of multipoint intragenic crosses. It was found that a considerarle proportion of double-site conversions can arise as two separate but correlated events. The same was true for conversion and crossing-over associated with it. The data point to a possibility of multiple recombinational events involving both conversions and cross-overs, occurring as successive rounds of recombination within a single recombinational process. Some of the results suggest that more than two chromatids may be involved in such a process. The results are discussed in terms of different recombination models.

If heterozygotes for a reciprocal translocation are intercrossed, some of their viable balanced progeny result from the fusion of unbalanced gametes with complementary duplications and deficiencies of the translocated segments. Therefore, if one parent in such an intercross is homozygous for a genetic marker on one of the segments concerned, some homozygous offspring will be produced even if the other parent does not have the marker. The expected frequency of such exceptional offspring among live-born is one-sixth if the marker is on the distal (non-centromeric) side of the point of exchange and single chiasmata normally occur in each interstitial segment. Much lower frequencies are expected if the marker is on the centromeric side, since duplications and deficiencies of proximal segments occur only as a consequence of adjacent-2 disjunction, in which homologous centromeres proceed to the same pole. This is rarer than normal disjunction. Thus, by comparing the frequencies of offspring homozygous for markers on one or other side of the point of exchange, it is possible (i) to determine which marker is in the centromeric segment, (ii) to estimate the frequency of adjacent-2 disjunction, given information on the nature of meiotic configurations in the translocation concerned.

By this method, it is shown that the frequency of adjacent-2 disjunction is similar in heterozygotes for mouse translocations (T5;18)26H, T(13; ?) 70H and T(14;17)264Ca, averaging 13%. Centromeres were located at the Sd end of linkage group V (confirming previous findings), the fz end of XIII and the bg end of XIV.

Coleoptile anthocyanin pigmentation in the hexaploid variety of wheat, Hope, is due to at least eight anthocyanins including four acylated forms. All are derivatives of either cyanidin or peonidin. Two homoeologous chromosomes, 7A and 7B, are involved in anthocyanin production. Both chromosomes carry genes that promote the synthesis of the same anthocyanins from flavonoid precursors. The roles of chromosomes 7A and 7B in anthocyanin biosynthesis and the consequences of interallelic interaction and dominance of possibly homoeologous loci are discussed.