This paper examines the hypothesis that the associations between alleles at allozyme loci and gene arrangements, which are observed in many Drosophila populations, result from inversions remaining associated with the alleles contained in the gametes in which they originally occurred. The effects of double crossing over in heterokaryotypes and of selection at loci linked to the allozyme loci, which are themselves assumed to be neutral, are studied theoretically. It is concluded that selection at linked loci is unlikely to have an important effect in retarding the decay of this type of association. The literature on associations between allozymes and gene arrangements is surveyed, and it is concluded that a ‘neutralist’ explanation of this sort cannot be excluded, except possibly in some cases where the allozyme locus is outside the inversion breakpoints and crosses over with measurable frequency in heterokaryotypes.

5-Bromodeoxyuridine (BrdU)-resistant mutants of Physarum polycephalum were isolated as colonies of myxamoebae growing on BrdU-substituted bacteria after exposure to long-wave ultraviolet light (UV). Twenty-four such mutants were studied. They all show Mendelian segregation in crosses with wild type. Plasmodia constructed from mutant amoebae were all deficient in deoxythymidine incorporation. Extracts made from selected plasmodia showed that all except one had low thymidine kinase activity.

Genetical and biochemical complementation studies revealed two complementation groups: 23 mutants, bur A, had low thymidine kinase while 1 mutant, bur B, had normal thymidine kinase levels.

A computer simulation of the genetic analysis of the asexual life cycle of Phycomyces' heterokaryons has been carried out. We have studied experimentally the relationship between the nuclear proportion in heterokaryotic mycelia containing prototrophic and auxotrophic nuclei and their growth rates. We discuss possible evolutionary implications of heterokaryosis in coenocytic fungi and the genetic applications of the quantitative complementation technique.

Growth of E. coli K12 in the presence of chlorpromazine or promethazine resulted in the appearance of a special kind of mutational change with a frequency of 5–10%. These cells were UV-sensitive and very mucoid in their colony formation. The strains were characterized as colanic acid producing on the basis of phage m-59 sensitivity and we tentatively conclude that these mutants are Lon−.

In another series of experiments drug treatment of E. coli failed to yield significant numbers of auxotrophs and the back mutation rate of a number of S. typhimurium His− strains was not significantly increased.

Gene frequency changes were followed in 17 experimental cage populations of Drosophila melanogaster polymorphic for the two loci G6PD and 6PGD. From these cages it was observed that selection was favouring the F alleles of both loci when the cages were started with low frequencies of these alleles (0·20F and 0·50F). A viability experiment and an extinction experiment also provided evidence that the F alleles of both loci were favoured. It is argued that the observed selection is not due to selective differences in genetic background of the different alleles, but is acting on the isozyme loci themselves.

The number of loci which are potentially able to produce sterility genes was estimated for Drosophila melanogaster. There appear to be, on the second chromosome, about 80 loci capable of producing male sterility and about 60 loci capable of producing female sterility. These figures seem to be considerably less than (400–500) loci responsible for lethal genes.

Thymineless strains of Escherichia coli C600 were constructred harbouring both an R factor of the N incompatibility group (R46 or R447b) and a compatible plasmid (Plac- of the A-C group or the Iα plasmid R62), which contained a segment of N group DNA. Selection was made for the transferred plasmid and dislodgement phenomena were manifest either as loss of an entire plasmid or as deletions of a region of plasmid DNA. Even after the two R factors had become established as separate replicons, the N group R factor but not the other plasmid exhibited instability.

Thymine starvation of strain C600 thy (R447b/R62) increased the elimination rate of the N group plasmid R447b but no elimination of R62 was observed. However, thymine starvation of strain C600 thy (R46/Plac-) not only increased the rate of elimination of R46 but also increased the rate of loss of Plac-. There was no detectable increase in nuclease activity in unstarved R46/Plac- strains and it is concluded that dislodgement of R46 from these strains is not due to induction of the nuclease that has been proposed to be responsible for the elimination of N group plasmids during thymine starvation.

Two variants of Plac- were isolated. These did not dislodge R46 from unstarved R46/Plac- strains and were not lost during thymine starvation even though thymineless elimination of R46 occurred at normal frequency.

An experimental evaluation of Robertson's (1970) theory concerning optimum intensities of selection for selection of varying durations has been carried out using published results from a long term selection study in Drosophila. Agreement of predicted rankings of treatments with expectations was excellent for low values of t/T (generations/total number scored) but poor for larger values of t/T. This was due to the 20% selection intensity treatments responding worse than expected and the 40% treatments relatively better than expected. Several possible reasons for the discrepancies exist but the most likely explanation is considered to be the greater reduction in effective population size due to selection in treatments with more intense selection.

The effectiveness of selection for increased pupa weight in Tribolium was compared for three different selection systems. In all three systems the same number of breeding individuals was used each generation. Population L was a large random mating population with 24 males and 48 females selected each generation. The C4 and C8 populations were each divided into 6 subpopulations (lines) consisting of 4 males and 8 fem ales. Each of the three populations was replicated. In C4, selection for pupa weight was within lines for three generations, followed by a generation of among-line selection when the best two out of six lines were selected. These lines were then crossed to produce 6 new subpopulations, and the cycle was repeated. The C8 population was handled in exactly the same manner except that seven generations of selection within lines were practised before each generation of among-line selection. Selection response for the 42-generation period was significantly greater in the L population than in either subdivided population. No consistent differences among the selection systems were apparent when evaluating short-term response for the first 12 generations of the experiment. The results were interpreted as indicating that the influence of multiple-peak epistasis was not of major importance for this trait in determining ultimate response to selection when starting from a base population of previously unselected lines and utilizing a within- and among-line selection regime.

The gene pfpr-10 in Coprinus confers resistance to p-fluorophenylalanine. The resistance of the heterozygote pfpr-10/pfp8-10 is completely recessive, both in diploids and dikaryons. A dominance modifier Mod+ makes pfpr-10 dominant in a dikaryon but not in a cliploid. It also enhances the degree of resistance in r homozygotes. These results are explained on a hypothetical model based upon a hexameric product of the pfpr-10 gene, with resistance to PFP being proportional to the percentage of homo r hexamers. The Mod+ is presumed to act by keeping the r and s products separate so that hybrid multimers are reduced to a minimum. Critical threshold concentrations of PFP for diploid/haploid dikaryons in 40 different combinations of pfpr and Mod+ genes cover a wide range of gene doses and percentage of homo r multimers. The relationship between the critical threshold concentrations and the calculated percentage of homo r multimers supports the model.

The development of a set of two-locus descent measures is reviewed. The three digenic measures, inbreeding coefficient and parental and recombinant descent coefficients, are considered in detail. The derivations of these three in pedigrees, fixed mating systems, and random mating in monoecious or dioecious populations are given. General expressions for digenic frequencies and disequilibria functions at any time are found by applying the three digenic descent measures to two types of initial populations. The final or equilibrium status of the population is also given. As the inbreeding coefficient is the same as the recombinant descent coefficient in the case of complete linkage, avoidance or promotion of early inbreeding has similar effects on the two coefficients. Estimable components of linkage disequilibrium and other measures of association within and among populations are elaborated.

Five biometrical traits (thorax length, wing length and width, sternopleural and abdominal chaetae numbers) were measured on 13 equatorial African strains and 30 French strains. In all cases highly significant differences were observed between the two geographic groups. These results are added to previously known variations concerning adult weight and ovariole number. In each place, the genetic particularities of the wild populations seem to be maintained by the selective pressure from environmental conditions, resulting in a homeostatic focusing of the best fitted average genotype. Analysis within each group showed that variations between strains were in most cases poorly or not correlated, so that partial or total genetic independence between the various traits measured seems likely.

The frequency of monogenic resistance to anticoagulant rodenticides in Rattus norvegicus in an area straddling the England–Wales border was monitored from 1967 until 1975. Rats were trapped on farms and tested in the laboratory by administering a dose of warfarin lethal to susceptibles. The mean incidence of resistance was 44% and did not change significantly, despite the extensive use of anticoagulants by farmers during the 9-year period. In 1975 more refined techniques showed that the frequencies of susceptible (SS) and resistant (RR) homozygotes were significantly below the Hardy–Weinberg expectations and simple estimates of the relative fitness ratios for the RR, RS and SS phenotypes were 0·37, 1·0 and 0·68 respectively. In two relatively isolated valleys, where selection with anticoagulants was minimal, the frequency of resistance decreased significantly from 57% to 39% during 1973–5. The results are consistent with the hypothesis that a balanced polymorphism is being maintained. Selection against susceptible homozygotes by the use of anticoagulant rodenticides, and against the resistant homozygote due to its high susceptibility to a primary deficiency of vitamin K gives the heterozygotes a selective advantage. A number of ecological factors that influence the incidence of the resistance are discussed briefly.

Thirty-nine enzyme loci and seven non-enzymic protein loci have been screened for electrophoretically detectable variation in a population of Pleuronectes platessa, a marine flatfish. The mean heterozygosity per individual is 0·102 ± 0·026, or 0·118 ± 0·030 if the non-enzymic proteins are excluded. The distributions of allele frequencies and single locus heterozygosities are given, and the results discussed with reference to current theories concerning the nature of protein variation.

Growth relations of lines selected for fast or slow larval feeding rate have been compared with those in the genetically heterogeneous control base population from which they were derived. Larvae of the slow strain have reduced growth rate and reach their critical weight for pupation later than unselected larvae. Larvae of the fast strain attain their critical weight at the same time as the unselected control larvae, suggesting that growth rate in the precritical period of development is already maximized in the base population and cannot be improved by increasing food intake. This constraint does not apply to the fixed period of post-critical growth however, since fast feeding larvae give rise to larger adult flies than the controls.

Larval feeding rate is affected by genes located on all three major chromosomes. The small fourth chromosome has negligible effect. Selection for slow feeding rate has led to an increase in the frequency of recessive genes affecting the character. High scores of larvae selected for fast feeding rate depend upon interactions between non-homologous selected chromosomes which individually have little effect. Larval feeding rate in the control unselected population appears to be buffered, firstly by epistatic interactions against the effects of chromosomes tending to promote ‘supra-optimal’ feeding rate and, secondly, by dominance against chromosomes promoting a lowering of feeding rate.

Under conditions of scramble type competition between the selected lines for limited resources, fast feeding larvae have a higher survival rate, and complete their period of larval development earlier to give larger adult flies than their slow feeding competitors. The contribution of larval feeding rate to competitive ability at different levels is discussed, and it is suggested that the effects of change in this behavioural character may be far reaching.

5-Bromouracil was shown to enhance the mutation frequency in an ultraviolet sensitive strain of Neurospora crassa (uvs-2) that was previously irradiated with ultraviolet light. The design of the experiment eliminated light or dark repair as possible hypotheses to account for the observed enhancement. Since a similar effect was noted at the rib-1 locus in N. crassa, which is not a UV-sensitive strain, it appears that the error prone repair mechanism in N. crassa may be related to post-replication repair. This enhancement may be attributed to the interference of 5-bromouracil with post-replication repair. It is also postulated that the maximum increase in the mutation frequency occurs when the greatest number of unrepaired pre-mutational UV lesions are repaired in the presence of 5-bromouracil.

Primary genes, designated as NYSA, NYSD, NYSE, NYSF (dominant) and nysB, nysC (recessive) are responsible for resistance to nystatin ranging from 30 to 80 u/ml. A dominant modifier gene increases the resistance conferred by NYSF from 80 to 140 u/ml and a recessive modifier gene enhances resistance due to NYSA by 20 u/ml (from 40 to 60 u/ml). One nystatin-resistant mutant is apparently cytoplasmic and this ‘mutation’ suppresses specifically the action of modifying factors which increase the level of resistance conferred by the NYSA gene. Interaction resulting in additive levels of resistance is shown by gene NYSD in combination with each of the genes nysB, nysC and NYSE. A model of step-wise increases in resistance due to polygenic primary genes and nuclear and cytoplasmic modifying factors is proposed.

The genetic properties of some new mutant t-haplotypes derived from the naturally occurring haplotype t6 have been investigated. Several mutant haplotypes gave taillessness with brachyury, T, were viable when homozygous, and when in compound with t6, and all these expressed tf and appeared to have arisen by meiotic crossing-over between T and tf in the region of crossover-suppression of t6. The haplotype th17 arose by crossing-over in a stock of t6 carrying the translocation T(1; 17) 190Ca, in which the translocation break in chromosome 17 is within the crossover-suppressing region; hence th17 could not be separated from T190. It was of the complementary crossover type in that it did not interact with T, and was lethal in compound with t6; its male segregation ratio was normal with heterogeneity. A further haplotype, th20, expressed tf, gave taillessness with T, and was lethal when homozygous and in compound with t6. The expression of tf was attributed to a deletion, which also covered the loci of the nearby lethals knobbly, Kb, and tw5. th20 showed weak complementation of tw5. The haplotype th7 had previously been interpreted as carrying a duplication; it was shown that this duplication was semi-lethal when homozygous. it is suggested that lethal mutant t-haplotypes commonly arise by duplication or deletion but such structural changes are not necessarily at the end of the haplotype, and need not arise by unequal meiotic crossing-over.

The R factors, R56 and R64, interfere with the growth of several otherwise unrelated phages, including λ, P22 and T7. In addition, λvir is inhibited more strongly in λ lysogens carrying either R factor when the prophage has an intact O – P region. Interference does not occur in certain str A R+ hosts but may reappear if they also carry other mutations affecting ribosomal proteins. Host mutants not permitting interference were isolated but were not strA. Non-interfering plasmid mutants were also isolated, of which some also failed to cause interference after transfer to a different host genus.

Large reciprocal differences in recombination frequencies were observed in F1 female progeny of a number of strain crosses. These reciprocal effects on recombination were found in several intervals of all three major chromosomes; they were greatest in intervals proximal to or spanning the centromere but were also found in some distal regions. The direction of recombination change was not consistent over different chromosomal intervals. There was no clear trend for reciprocal recombination differences to be associated with change of interference values between adjacent intervals. A close association was found between reciprocal female recombination effects, male recombination and other component traits of hybrid dysgenesis. However, reciprocal differences in female recombination were not restricted to dysgenic crosses. Backcross experiments demonstrated that reciprocal differences in the centromeric region of chromosome III were the result of increased crossing-over in dysgenic F1 hybrids rather than decreased crossing-over in the reciprocal hybrids. It is concluded that genotype-cytoplasm interaction can be a major factor influencing recombination frequencies and that the use of interstrain hybrids for recombination measurement may lead to errors in the estimation of intrastrain frequencies.