Segregations at five loci on the left arm of chromosome VII and other data from published sources were used to test a new mathematical model of meiotic segregation in trisomics of the yeast Saccharomyces cerevisiae. Results support this model which predicts that non-centromere-linked genes in trivalent complexes segregate at random. The data indicate that bivalent–univalent complexes are infrequent in trisomics and that recombination between all three homologues is frequent in regions close to the centromere. A test of homologue interference produced no evidence that a crossover between two homologues has any influence on the probability that the same two homologues will be involved in an adjacent crossover.

Transformation studies in pneumococcus had shown that loci determining the resistance to erythromycin and streptomycin were unlinked when strain Cl3 was recipient but linked when strain SIII-I was recipient. This phenomenon also applies to other pairs of markers studied in these two recipients, no matter whether the transforming DNA was derived from strain Cl3 or strain SIII-I. Other differences between the two recipient strains were also revealed. Whereas competent cultures of strain Cl3 were composed of all competent cells, which was in agreement with previous reports of pneumococcal cultures, strain SIII-I normally gave a maximum average of 28% competent cells. Strain SIII-I was unstable, since on repeated sub-culturing the competence peak profile changed and the value of ‘fcq’ increased. These properties were reflected in the two hex− strains 401 and R6x which were found to be similar to the ‘altered SIII-I’ strain. The results from the linkage studies have been applied to the chromosome map and have placed the opt-r2 gene in the str-r41-containing chromosome arm.

Klebsiella aerogenes V9A carrying a lac plasmid in addition to its chromosomal operon showed strongly positive fermentation of lactose on MacConkey lactose agar plates, and was found to transport the lactose analogue thiomethyl-β-galactoside (TMG) at a rapid rate. The strain that had been freed of the plasmid showed moderate transport due to the chromosomal lac operon. When a plasmid bearing a mutation in lac Y was inserted into a strain with a normal Y gene, the resulting diploid became lactose-negative in phenotype. The presence of E. coli F′lac factors that carried lac Y mutations, whether deletions or missense or nonsense mutations, also rendered lac Y+Klebsiella lactose-negative. Such diploids, after growth in 1% lactose, transported TMG at a much lower rate than the corresponding plasmid-free lac Y+Klebsiella. However, this interference by lac Y− plasmids with the expression of the chromosomal lac Y gene was not seen when cells were induced with IPTG or when the chromosomal and plasmid lac operons were both constitutive. It was found that this effect of the plasmids was dependent on their possessing an intact lacZ gene.

Two systems of hybrid dysgenesis, the P–M system and the I–R system, are characterized by two different specific types of non-reciprocal hybrid sterility referred to, respectively, as GD and SF sterility. In order to determine the relationship between these two systems, strains representative of the four single-system interactive types were crossed in almost all possible combinations and tested for both types of sterility. The results suggest that the two systems are at least partially independent of one another. There are several examples of single strains that contribute the maternal component for interaction in the P–M system and contribute the paternal component in the I–R system. Using parents with the potential for the two types of interaction and. appropriate temperature manipulation, both GD and SF sterility can be manifested in the same hybrid females. In other crosses, a single type of sterility is observed, or none at all, according to the dual designation of the parental strains. The evidence from a number of additional crosses suggests that most strains of this species have the potential for both types of interaction to varying degrees when mated in appropriate combinations. Some theoretical and practical implications of these results are discussed.

Populations of linear asci are classified according to the number of spores in an ascus. The resultant five numerical classes are further classified into ten spatial classes according to the arrangement of the spores in an ascus and, by inference, into ten historical classes according to the number and origins of failures during the developmental process. An analysis of the observed frequencies of numerical classes allows derivation of the efficiencies of the first meiotic division, the second meiotic division and sporulation in a fission yeast. The analytical method may be useful in locating the site of action of sporulation inhibitors and in identifying meiosis mutants from sporulation mutants.

From the F2 ratios of crosses between Drosophila melanogaster strains homozygous for the AdhF(F) and the Adhs(S) allele it has been concluded that the developmental time of FF homozygotes is shorter than that of SS homozygotes. This difference is found to be reinforced by increasing levels of crowding. A further analysis of developmental times has been performed by the transfer of larvae to agar medium after they have stayed for periods of variable length on regular food. From the percentage of emerging adults it can be concluded that FF larvae and, to a lesser extent FS larvae, either reach their critical weights for pupation earlier than SS larvae or possess a lower critical weight. These differences in developmental time influence the course of allele frequencies. Between populations kept on a 2-week transfer schedule and on a 3-week schedule a divergence of allele frequencies is observed, in the former a decrease in S frequency occurs. The relevance of the observed differences in developmental time for the maintenance of the Adh polymorphism is discussed for laboratory populations kept on regular food and at varying densities.

Hybrid dysgenesis has been described as a syndrome of aberrant traits including sterility, male recombination, and mutation, which occurs in some inter-strain hybrids of Drosophila, but only from one of the two reciprocal crosses. In a series of experiments in which hybrids of various pedigrees were tested for sterility, it was found that a case of hybrid dysgenesis could be most easily interpreted as the interaction of two components. One component was found to be a polygenic Mendelian factor linked to each of the major chromosomes of π2, the paternally contributing strain (‘P strain’). These chromosomes were capable of causing sterility when inherited from either parent, provided the appropriate maternal component was also inherited. The ability to transmit this maternal component was designated ‘cytotype’ to indicate that it is a property of the entire cell. It was possible to classify nearly all hybrid females as either P or M cytotype on the basis of their ability to produce sterile daughters. All daughters of the M-cytotype mothers were susceptible to the sterilizing effects of the π2 chromosome, whereas all, or nearly all daughters of P-cytotype mothers were immune. When more than one of the π2 chromosomes were received by daughters of M-cytotype females, chromosomal interactions could be detected statistically, but the model of independent action remained a useful approximation. Cytotype was shown to be determined by chromosomal factors, but with limited cytoplasmic transmission. This unusual mode of inheritance can be compared with other cases of hybrid dysgenesis where the behaviour resembles that of self-replicating cytoplasmic particles which are dependent on certain chromosomes. The lack of sterility from intra-strain crosses can be explained by the fact that chromosomes capable of causing sterility also induce the P cytotype, and thus prevent sterility in the next generation.

Deterministic simulations have been carried out of populations under normalizing and disruptive selection for a trait determined by genes with additive effects at six loci. In some simulations a pair of alleles at a seventh locus determined the rate of recombination between the seven loci. Normalizing selection with a single optimum, fixed or fluctuating, invariably led to genetic homozygosity. If the optimum fluctuates widely, the approach to homozygosity may be accompanied by a large decline in the mean fitness of the population. Disruptive selection was simulated by having two ‘niches’ with separate optima and separate density-dependent regulation, but with the adult population mating randomly. If the optima are widely separated, this leads to stable polymorphism. Selection produced linkage disequilibrium, normalizing selection causing repulsion and disruptive selection coupling between + and − alleles. This linkage disequilibrium accelerates the phenotypic response to selection, but delays changes in gene frequency. Selection always favoured alleles for low recombination at the expense of alleles for high recombination.

The effects of finite population size, occurring either as a bottleneck in a single generation followed by a large expansion or in all generations, are considered for models of two linked heterotic loci. Linkage is assumed to be tight because it is required if there is to be stable linkage disequilibrium, D ǂ 0, in infinitely large populations. (D is the difference between gamete frequencies and the product of the gene frequencies.)

If a substantial perturbation of frequencies occurs as a result of a bottleneck but the population is subsequently very large, D may take hundreds of generations to return to its stable point. In finite populations, the distribution of D can be ⋃-shaped, unimodal or bimodal. The correlation of D in successive generations is higher with tight linkage and is little affected by selection or the size of the population.

The utility of infinite population studies of linkage disequilibrium and its stable points is questioned, and considerable pessimism is expressed about the possibilities of distinguishing selection and sampling effects at linked loci.

There is a strong genetic component to fertility differentials among individuals of ponderosa pine. Prolific cone-producers as a group were markedly different from low cone-producers at the three protein loci which were monitored. The two groups did not differ significantly in age, but trees with high cone production had slower growth rates and smaller diameters than trees with low cone production. To our knowledge, these results provide the first demonstration of fertility differentials associated with specific genes in a woody plant.

Spermatozoa from inbred strains of mice were found to vary significantly for levels of cyclic AMP when extractions were performed in a reproducible manner. The F1 hybrid between high and low spermatozoal cAMP strains showed spermatozoal cAMP levels typical of the low strain. An analysis of spermatozoal cAMP in individual mice from the back-cross of the F1 to the high strain suggested that alleles at more than one locus determine strain differences in spermatozoal cAMP. The major histocompatibility locus of mice, H-2, which had been found to have an effect on liver cAMP levels did not seem to affect spermatozoal cAMP levels. t-Alleles, which appear to alter fertilization rates by effects on motility, had no apparent affects on spermatozoal cAMP.

Escherichia coli K-12 mutants resistant to glutamyl-γ-methyl ester were isolated. A mutation leading to resistance of up to 1·4 mg/ml of the methionine analogue maps at min 63 and is 13% cotransducible with serA indicating an alteration in the metK gene. Another mutation leading to resistance to 3 mg/ml of the analogue and cross-resistance to other amino acid analogues maps at min 87. This mutation, which has the phenotype of MetJ−, is shown to be situated between the glpK and metB genes and thus indicates a different gene order from the published one.

A new autosomal recessive gene (cab, ‘cardiac abnormality’), which affects heart differentiation appeared spontaneously in our mouse colony. The affected homozygotes die approximately half a day prior to birth or at term, with the myocardium exhibiting vacuolation and large areas of mesenchyme-like cells. A significant finding in mutant foetuses is the lack of histochemically detectable glycogen.

Male sterility, male recombination, and transmission ratio distortion – all examples of a syndrome known as hybrid dysgenesis in Drosophila melanogaster – were found to involve chromosome–cytoplasm interactions. The latter two have temperature optima near 25° and involve pre-meiotic events. In addition, sex ratio distortion, and induction of certain translocations of the X and Y chromosomes (but not the autosomes) were found to be part of hybrid dysgenesis. Both are caused by chromosome–cytoplasm interactions with pre-meiotic events playing a crucial role. The results agree with previous data on female sterility in hybrid dysgenesis, which also has cytoplasmic components and premeiotic origins.

This paper tests the hypothesis that haplodiploidy or X linkage leads to less genetic variability. Although haplodiploid organisms exhibit a low level of genetic variability the wide variation existing between different diploid organisms implies that factors other than the genetical system could also be responsible. In order to test the hypothesis critically it is necessary to compare the level of genetic variability between X-linked and autosomal genes within a closely related group of organisms. For kangaroos, the ascertainment bias for X-linked loci has been removed by assuming the correctness of Ohno's law of conservation of the mammalian X, i.e. that genes found to be X-linked in man can be assumed to be X-linked in kangaroos. For Man and Drosophila, it has been assumed that the percentage of the karyotype which is X chromosome can be used as the expectation for the percentage of X-linked polymorphisms. No difference between the two classes of loci is evident in kangaroos and man for percentage polymorphism. The data however have confidence limits which would allow autosomal loci to have three times greater percentage polymorphism. In Drosophila the published data of Prakash show that autosomal loci are polymorphic about twice as frequently as are their X-linked counterparts. Thus there may be a modest reduction in percentage polymorphism as a result of X-linkage (i.e. haplodiploidy). No reduction in the number of alleles per locus or average heterozygosity at those loci which are polymorphic is evident in kangaroos, man, or Drosophila. More data on more X-linked enzymes are necessary to establish firmly that there is a real reduction in percentage polymorphism and to estimate its extent. The kangaroo data are incompatible with the hypothesis that a large fraction of the variability is maintained by simple overdominance since overdominance is very unlikely in the quasi-haploid genetical system which results from the paternal X inactivation mode of dosage compensation used by kangaroos. This is the first report on level of enzymic variability in marsupials. 17% of autosomal loci and 18% of X-linked loci are polymorphic, average heterozygosity is 4% for autosomal and 4% for X-linked loci and number of alleles per locus is 1·25 for autosomal and 1·21 for X-linked loci. These figures are somewhat lower than for eutherian mammals.

Strains of Escherichia coli harbouring a conjugative plasmid that carries a transposon for tetracycline resistance (Tn1771) were found to be adaptable to very high tetracycline concentrations. The molecular analysis of plasmids isolated from strains with enhanced levels of tetracycline resistance revealed an intramolecular amplification of the resistance determinant of the tetracycline transposon.

A model for the molecular structure of the transposon is presented, which suggests that there are three repetitive DNA segments on Tn1771. This accounts for the properties both of transposition and of gene amplification.

We calculated the rates of segregation due to plasmid incompatibility under several simple models. A common feature of all the models that we considered is that incompatibility is caused by the inability of the segregation mechanism to distinguish between two incompatible plasmids.

We measured the rate of segregation due to incompatibility of a pair of ColE1 derivatives under two conditions: (1) One plasmid was introduced into cells carrying the other by conjugation. (2) Cells carrying both plasmids were maintained by selection and then selection was released.

Interpretation of the results was made more difficult by effects of the Plasmids on the host cell's growth rate. These experiments gave results in agreement with the predictions of a random pool replication model. Published results were also in reasonable agreement with this model.

An electrophoretic variation in a mouse serum protein moving closely to the vitamin D binding protein is described. The variation is determined by two codominant alleles at one locus with the allele in DBA causing fast mobility and that in C57BL causing slow mobility. This locus is located in the proximal part of chromosome 9, 14·3 cM from the d locus and 31·9 cM from the Bgs locus. The protein has not yet been identified but possible candidates among the serum proteins are discussed.

Methionine-sensitive mutants of Schizosaccharomyces pombe were found to map at 16 loci, min1–min16, seven of which are allelic to known ade loci involved in adenine biosynthesis. The mutant min1–106 was shown to be located within the ade6 locus by intragenic recombination analysis. Methionine sensitivity of these mutants is reversible by adenine supplementation and their pigmentation is similar to that of the corresponding ade mutants. We conclude, therefore, that these min mutants are incompletely blocked in adenine biosynthesis. These observations further indicate that adenine biosynthesis in S. pombe is inhibited by methionine or a metabolite derived from it. For some other min mutants showing reversibility of methionine sensitivity by leucine or proline, no allelism to known leu or pro loci was detected.

A method is introduced that allows the simplification of the calculation of equilibrium solutions in multiple locus genetic models of a single infinite population. The method can be applied when the number of different fitnesses is equal to or less than one more than the number of independent allelic frequencies. The results are in terms of relationships – the symmetry constraints – between the gametic frequencies that must be satisfied at any boundary or internal equilibrium. The symmetry constraints are independent of the fitness values and of the recombination fractions. This can lead to some understanding of the equilibrium structure of a model when the full equilibrium solution is not obtained and reduces the number of independent variables in the calculations of the full equilibrium solutions. Examples of two locus models with two alleles at each locus and with two alleles at one locus and three at the other are discussed.