1. Dominant hemimelia (Dh) and postaxial hemimelia (px) both belong to the luxoid group of mouse mutants, tending to cause luxation of limb-bones associated with hemimelia and polydactyly or oligodactyly.

2. Dh is about 4·8 units from leaden in linkage group XIII, while px is about 2·9 units from the microphthalmia locus in linkage group XI.

3. The nature and range of skeletal anomalies found in Dh heterozygotes closely mimic those found in luxate heterozygotes and homozygotes. As in luxate, only the preaxial side of the hind-limb is affected, with (i) polydactyly or oligodactyly (sometimes also syndactyly), (ii) tibial hemimelia, (iii) reduction and fragmentation of the femur, (iv) reduction of the pubis. The more proximal the anomaly the less frequently does it occur; about 4% of Dh/ + mice show no limb anomaly. The defects in Dh homozygotes are similar but usually more extreme, with severe oligodactyly, loss of tibia, fragmentation of femur and reduction of pubis.

4. All mice carrying Dh lack the spleen. Stomach size is reduced slightly in Dh/ + and greatly in Dh/Dh mice. In Dh/ + mice, the left kidney is either flattened antero-ventrally or (less frequently) hydropic; the right seems normal. Dh/Dh mice nearly always have severe hydronephrosis, as well as posterior visceral defects similar to those of the ‘uro-recto-caudal syndrome’. They usually die before weaning, but a few survive to maturity and a female has even bred.

5. Preliminary developmental studies show that Dh/Dh skeletal defects can be traced back to the precartilaginous stage.

6. When homozygous, px affects the postaxial side of the fore-limb and sometimes of the hind-limb also; a large ‘foramen infraspinatum’ is always present in the scapula. There may also be (i) oligodactyly and occasional syndactyly, (ii) ulnar hemimelia, (iii) distal reduction and distortion of the femur. Fore-limbs tend to be more severely affected on the right, but hind-limbs on the left. Extra sesamoid bones occur on the extensor side of digits in all four feet and are associated with extra tendons.

7. Both sexes are sterile and show anomalies of the paramesonephric (Müllerian) ducts. In the female, the vagina is wholly or partly double and the uterine tubes uncoiled, with an abnormal relationship to the ovary. In the male the paramesonephric duct persists in the adult, while the vas deferens runs into the distal end of the vesicular gland instead of into the urethra. The epidermis of manus and pes shows abnormal dark papillae on the extensor side.

8. The mean number of presacral vertebrae is reduced by 0·8 in Dh/ + and 1·6 in Dh/Dh mice. It is also reduced in high-grade postaxial hemimelics.

9. The relationships of these two genes to other luxoid mutants are discussed.

1. It has been shown that the lines dp 1, dp 2, vg 4 and vg 6 of Thoday & Boam (1961) each have two high sternopleural chaeta number genes or ‘effective factors’ between h and eyg in chromosome III. Their line dp 6 does not contain these two genes.

2. Lines derived from ancestors of dp 2 and vg 4 before the latter produced their accelerated responses have third chromosomes affecting chaeta number as if they had only one or other of these genes.

3. Of the three stocks from which all the lines derived, one, Inbred Oregon, lacks these genes. The second, vg/vg, has third chromosomes similar in effect to Oregon. The third, dp/dp, was heterogeneous, having a class of third chromosomes similar in effect to those of Oregon and a class similar to those having one high gene.

4. It is suggested that the history of the accelerated response in dp 1, dp 2 and vg 4 was as follows. Initially most of these third chromosomes were − − at the two loci, but a minority (derived from the dp/dp stock) were + − and − + (where + indicates the allele increasing chaeta number. Selection would reduce the frequency of − −, and hence increase the proportion of + −/− + heterozygotes and the probability of recombination to produce + +. Origin and multiplication of + + would account for the accelerated response.

An experiment was conducted with the flour beetle, Tribolium castaneum to investigate genotype-environment interaction. Ninety-two matings of one male and three females were made at random from a large panmictic population. The females were transferred to individual containers after 4 days. Six daughters were randomly selected from each sire–dam pair and individually mated to unrelated males. Three of the six were placed in an incubator (33·3°C., 45% relative humidity) and three in a cabinet at room conditions (22·2–26·7°C., 30–35% relative humidity) and allowed to produce eggs during a 3-day period. Progeny were counted as pupae and larvae. The traits studied were number of pupae and number of pupae plus number of larvae. A transformation to (X +1)½ was required. A conventional least-squares model was employed, and a large environmental effect was observed. In the incubator the mean number of pupae was 13·4 and of pupae + larvae was 20·1 while the corresponding figures for room conditions were 3·3 and 9·2. Genotype-environment interaction accounted for 3·7 to 6·7% of the total variance for (pupae + 1)½ and 2·1 to 8·3% for (pupae + larvae+ 1)½. Heritability of the traits was essentially the same in both environments. The interaction was due to an increasing difference between environments in production associated with increasing breeding values of the sire, and to small changes in rank of breeding values on the two environments. As a result of the interaction, selection in one environment for production on the other would be expected to be only 71 to 72% as effective as direct selection for (pupae + 1)½ and 62 to 86% for (pupae + larvae + 1)½ even though the fraction of the total variance attributed to genotype-environmental interaction was less than 10%.

The short-term response to artificial selection in a wild-type population for increased scutellar bristle number, has been interpreted in terms of two alternative threshold models. Both models have been shown to give a satisfactory transformation of the scale of measurement, in that the underlying variable shows an effectively linear response in terms of the accumulated selection differential, and the pheno-typic variance on the underlying scale remains virtually unchanged over a period of generations sufficient to increase the incidence of extra-bristles to almost 100%.

A genetic analysis based on the bivariate model, which to a large extent takes account of the location of the supernumerary bristles, has shown the phenotypic variance on the underlying scale to have the following composition: the additive genetic variance is 27·2 ± 0·9% of the total, and the developmental error or ‘chance’ variance accounts for a further 63·8 ± 2·6%, so that the remaining 9·0 ± 2·7% of the variance must be due to environmental differences among individuals and to non-additive genetic effects.

The relative magnitudes of these parameters are remarkably similar to those observed for two quasi-continuous variables in the same population, viz. abdominal and sternopleural hair number. In view of the ease with which scutellar bristles may be scored, the character can therefore serve as an extremely useful model character for quantitative genetic experiments.

Genetic consequences of non-random segregation of chromosomes are discussed. In the experimental part preferential segregation of two linkage groups in Ascobolus immersus is described. This was established in several crosses involving mutants 164, XXVI and 231 (all in the same linkage group) and mutant 726.

The selection of animals or plants for high values of a certain character may favour not only genotypes associated with these high values but also genotypes associated with high variability. Any differences between genotypes in variability may therefore be of considerable importance in plant and livestock improvement programmes as well as in evolution. The effects of various selection procedures on variability have been studied in three recent experiments [Falconer & Robertson (1956) Falconer (1957) and Prout (1962)]. In these experiments one line was continued by selecting, in each generation, parents with values of a particular character near the population mean. Manning (1955, 1956) has described the effects of this kind of selection applied to cotton. Robertson (1956) derived and discussed the theory of such selection procedures when certain simplifying approximations can be made We shall obtain some more general results and show that Robertson was incorrect in saying that the selection procedure would lead to gene fixation even if the heterozygotes are less variable than the homozygotes. The importance of the results is discussed in section 8.

1. Chromosomal polymorphism in two populations of D. subobscura from Barcelona has been analysed and compared with the polymorphism which has been reported from other populations.

2. In spite of microclimatic differences between the Barcelona sites, there is no evidence of differences in the frequency of inversion types nor of seasonal differences within one of these sites.

3. Regularities in the array of frequencies of chromosomal types in Barcelona and other populations suggest that the polymorphism is mainly adaptive although there is also evidence of differences due to distance which are not obviously associated with differences in habitat.

4. In general the index of free recombination gives higher values in the marginal populations of D. subobscura, but further data are needed to arrive at a definite conclusion on this point since the present evidence is rather ambiguous.

1. Two total and five leaky sorbitol mutants isolated in Aspergillus nidulans by defective growth on the sugar are all recessive. The mutants are closely linked, they appear to represent three linked genes spanned by a deletion.

2. Mutants which complement in heterozygous diploids do not complement in balanced heterokaryons. Failure to complement is a property of the mutants and not the result of a nutritional interaction or an unfavourable nuclear ratio in the heterokaryons.

3. Sorbitol is oxidized by an inducible enzyme system in the wild-type. There are at least two enzymes concerned in the oxidative assimilation of sorbitol, an initial oxidative enzyme, which is defective in the leaky mutants, and a later enzyme defective in the total mutants. There may also be a second non-oxidative pathway for sorbitol metabolism.

4. In diploids complementary pairs of mutants oxidized sorbitol at 75% the rate of the wild-type but non-complementary mutants did not oxidize the sugar. In balanced heterokaryons none of the pairs of mutants oxidized the substrate. It is concluded that failure of inter-genic complementation in the heterokaryons is the result of a failure of either enzyme formation or enzyme function. Models to account for differences in enzyme formation in heterokaryons and diploids are suggested.

1. Inactivation of one X chromosome in somatic cells of female mammals is a form of dosage compensation of sex-linked genes, but the mechanism is entirely different from that operating in Drosophila. The latter is designated as dosage compensation sensu strictu.

2. There is no dosage compensation of barred, sex-linked dilution or slow-feathering in domestic fowls, of almond or faded in pigeons, or of cinnamon in canaries. Among Lepidoptera the same is true of sex-linked melanism in Lymantria monacha and of a locus controlling haemolymph colour in Choritoneura spp. There is no positive evidence that dosage compensation occurs outside Drosophila and mammals.

3. Sex-chromatin in female birds (heterogametic) has been reported by several authors; the genetical evidence is against the possibility that this represents (as in mammals) an inactivated X chromosome. Sex-chromatin in the heterogametic sex also occurs in some (not all) Lepidoptera and Heteroptera; in Heteroptera it usually represents a heteropyknotic Y chromosome.

4. Some complications in Muller's theory of dosage compensation sensu strictu are discussed. Not all ‘compensatory modifiers’ are necessarily sex-linked.

5. The problem of dosage compensation in species with impaternate males is discussed; fused in Habrobracon is not compensated.

A selection of single and double mutant strains was backcrossed repetitively at least four times to each of the wild-types Abbott 4A, Abbott 12a and Lindegren 1A, and then re-isolated. Twelve crisp progeny isolated from the fifth backcross to Abbott 4A were heterogeneous for recombination frequency in the interval between cr and me-6 while thirteen crisp progeny from the fifth backcross to Lindegren 1A were homogeneous for recombination frequency in this interval. Thirteen crisp strains from the fifth backcross to Abbott 12a were homogeneous for recombination frequency between cr and nic-1 loci. However, mutant strains which had been backcrossed independently to Lindegren 1A were heterogeneous for recombination frequency in the intervals cr-aur and aur-nic-1.

In general, recombination frequencies in crosses between strains of the same wild-type ancestry including Abbott 4 and Abbott 12 were significantly higher than those in crosses of Lindegren × Abbott 4 or Lindegren × Abbott 12 ancestry but exceptions were found. Recombination frequencies between the same markers usually, but not always, increased on inbreeding and the changes in frequency were non-uniform in the marked region in some crosses.

The limitations of backcrossing as a means of transferring a mutant marker into a wild-type genome were discussed. It was concluded that inducing marker mutations anew in a given wild strain, preferably Lindegren 1A, might be more successful in obtaining constancy of map distance.

The genetics of two polymorphic loci in the fowl was studied. Each affects the electrophoretic mobility of a different egg albumen protein, and at each locus two alleles were shown to segregate in a Mendelian manner.

The polymorphic proteins are not any of the well-characterized egg albumen proteins, viz. ovalbumin, ovomucoid, ovomucin, flavoprotein, avidin and lysozyme.

In contrast to all others, two hens were found to have an anomalous, variable, phenotype. A tentative explanation of this was offered.

Lysis of stock 7 killers with ½% sodium deoxycholate, yields preparations with one half as much killing activity as homogenates, while lysing virtually all of the bright particles known to have killing activity in homogenates. The killing activity which persists is associated with particles which sediment at the same rate as the refractile bodies, the only visible remnant of the bright particles. It is concluded that killing activity is probably associated primarily, if not exclusively, with the refractile bodies of the bright kappa particles. Partially purified killer particles when stored at 27° C. for more than one day, progressively lose their ability to produce spinning of sensitives but retain their ability to produce paralysis. Loss of activity in crude preparations of refractile bodies kept 2 hours at 31° C. can be retarded by homogenates of either killers or sensitives.

1. The problem of whether or not hybrids are more efficient, less variable and have less stringent nutritional needs than their parents is examined by finding the dose-responses of four inbred Drosophila lines and their crosses to casein, choline, RNA, riboflavine, nicotinic acid, pyridoxine and folic acid, under germ-free conditions.

2. Under more or less optimal conditions, survival, development rate and weight of the hybrids are generally superior to those of the inbreds; the 4-way cross is not inferior to the 2-way crosses. Variability of the crosses is not necessarily lower than that of the inbreds, and the 4-way cross is no more variable than the 2-way crosses.

3. As measured by growth rate, the hybrids use casein more effectively than the pure lines, but their relative efficiency declines as the casein supply is decreased. There is a positive correlation between casein requirements for optimal growth and minimum requirements of pyridoxine. Hybrids also tend to be more efficient in their use of choline, but not of the other nutrients examined.

4. Deficiencies of particular nutrients (and also of excess provision of the non-vitamins) affect the lines and crosses differently, so that their relationships to one another are altered. The hybrids show no special advantage in resisting departures from the optimum. Variability is also changed significantly under sub-optimal conditions and, in some situations, the hybrids may then be more variable than the inbreds.

5. Each line and each cross is found to have its own optimal nutritional environment, and its particular reactions to departure from this. The full potential of the genotypes cannot be manifest, therefore, by tests in a single, standard environment.

Five dimensions—wing length and width, thorax length, head width and tibia length—have been measured on samples of twelve populations of Drosophila subobscura taken from different parts of the species range, extending from Scotland to Israel. The populations had been started from thirty or more pairs of flies and maintained in the laboratory for eight to eleven generations. They were reared for measurement under standard conditions, so that any differences between them must be of genetic origin. The localities from which the populations came ranged over 25° of latitude and 15° F. in mean annual temperature.

The population means of the five dimensions all showed strong positive correlations with latitude and slightly lower negative correlations with mean annual temperature. There is, in consequence, a very uniform cline in the five dimensions, each increasing as we move northwards. This pattern differs from that found for essentially the same dimensions of D. robusta by Stalker & Carson (1947), where all the correlations with latitude are lower and those for head and thorax size are of opposite sign in the two species.

A partial correlation analysis and reference to selection experiments suggests that there are two distinct groups of genes involved in these clines: (1) a group of genes causing increase in relative wing and leg size, and responsible for the positive correlation of these dimensions with latitude in both species, and (2) a group of general size genes causing correlated changes in all dimensions, which have been selected in opposite directions in the two species, with the result that head and thorax size increase in subobscura but decrease in robusta as latitude increases. The ratio of wing length to thorax length has a high positive correlation with latitude in both species.

Further analysis of the data available on both species brings out the following points:

1. In subobscura there is considerable variation between localities remaining for all dimensions when the correlations with latitude are eliminated. This component of variance is much larger in proportion to the within-locality variance in subobscura than in robusta.

2. Comparison of the percentage regression coefficients of dimensions on latitude or temperature shows that individual dimensions change more rapidly in proportion to both geographical indices in subobscura, but the ratio of wing to thorax length ?changes more rapidly in robusta.

3. The ‘environmental’ regression of dimension on temperature of rearing in the laboratory, measured only for robusta, is very much higher than the corresponding genetic regression between populations on mean annual temperature for all the robusta dimensions and different in sign for head and thorax size. But this environ mental regression is uniformly only about twice as high as the corresponding genetic regression for each character in subobscura. Unexpectedly, the environmental and genetic regressions on temperature of the wing/thorax length ratio are equal in robusta.

4. D. subobscura is about 20% smaller in head and thorax size, and 26% less in wing width, than robusta, when both species are reared at 18° C.

5. The basic within-culture variances of the five dimensions are 50–100% greater in subobscura than in robusta for all dimensions. Variability in D. melanogaster is about the same as in D. robusta.

Some of the possible implications of these results are discussed.

The pleiotropic effect of the rose-comb gene (R) on fertility when combined with artificial selection against the single-comb type (r) results in an interesting example of operational over-dominance. An evaluation of the equilibrium frequency of the rose-comb gene based on this over-dominance concept provides a plausible explanation for the relatively high frequency of single-comb birds appearing in the Wyandotte breed of fowls.

1. Ovalbumin was shown, by starch gel electrophoresis, to exist in two genetically different forms, A and B. It was suggested that these are determined by two alleles at one locus, named Ov.

2. Ovalbumin of each genetic type is electrophoretically heterogeneous. Dephos-phorylation of each type with human prostatic phosphatase and calf intestinal phosphatase removed some of the heterogeneity, but some remained.

3. The genetic difference was shown not to reside in the fragment released from ovalbumin by the proteolytic enzyme subtilisin.

4. The genetic difference was evident in the ovalbumin present in the fluid contained in right oviduct cysts.

This paper presents the first part of a further analysis of data collected many years ago by Clark and Hersh on allometric growth in Notonecta undulata, the water boatman. The data consist of measurements of a number of dimensions, taken each instar, on 72 individuals reared separately from egg to adult. These individuals were the progeny of nine wild females, so that a rough estimate can be made of the magnitude of genetic effects by comparing the variances within and between families. The paper examines the allometric relationship between femur length of the three legs and body length.

The statistical problems in this type of analysis are discussed. There are significant deviations from linearity when the six points for femur length are plotted against body length on a double log graph, but these are small compared with the overall linear trend. The growth coefficients (k) are calculated for the three femurs against body length for each individual, and are analysed for differences between femurs, between sexes and between and within families. This analysis shows that there are significant differences between the three femurs in growth coefficient, but the growth gradient, or pattern of k values between the three femurs does not vary significantly from one individual to another. On the other hand, mean k for the three femurs shows significant individual variation and also a significant variation between families. From this it appears that about one third of the phenotypic variance in mean relative growth rate of the femurs is due to genetic effects.

A computer has been used to investigate the effect of an initial period of reverse selection on the subsequent response of a population to renewed forward selection with the same population size and selection intensity. As the computer was used to derive gene frequency distributions, there was no random element in the results obtained. A theoretical solution to the problem was obtained for genes with small effects.

The process can be adequately described by the duration of the reverse selection (expressed in terms of the population size N), the product of population size and gene effect, Ns, and the initial gene frequency. If the duration of reverse selection, t, is less than N/2, the loss in selection advance due to the reverse selection is roughly t/N, though slightly greater than this for genes with low frequency. The ‘point of no return’ after which it is impossible, with the same population size and selection intensity, to return even to the starting frequency is 1·4N generations for genes with small effect and this declines as the gene effect increases.

Some extension of results to recessive genes is also given.

Cultivated plants of Nicotiana alata are self-incompatible and are of two kinds: normal (N); and exceptional (M). N plants are reciprocally compatible with N. langsdorffii; M plants are compatible only as males. M plants contain an unusual allele, SFI, which has a dual action in the style: it rejects both self-pollen, and Sf pollen from N. langsdorffii. The overall results agree with the assumption that the SFI gene produces two kinds of specificity in the style: primary specificity, which is responsible for the rejection of Sf pollen; and secondary specificity, which is responsible for the rejection of self-pollen as in SI alleles generally. The genetic sub-units concerned must be closely linked; there was no evidence for their dissociation in the 599 plants studied.

In both compatible and incompatible pollinations, SFI pollen grows more slowly than SI and, in addition, appears to depress the normal rate of growth of SI pollen. In consequence, crosses SfSf × SISFI ♂ yielded significantly fewer S.I. plants than the 50% expected. The two kinds of pollen grew at comparable rates, however, when F1 (M × M) plants involving parents from different original sources were backcrossed to SfSf ♀. Progenies then showed the expected 1:1 ratio of S.I. to S.C. plants. These results are assumed to be due to differential behaviour of the SFI allele according to its genetic background. The change in background would be from a degree of homozygosity, in plants from the same source, to a degree of heterozygosity, in crosses between plants from different sources.

The high incidence of the SFI gene in N. alata is considered to be due to the advantage it confers on a self-incompatible population when it is overlapping with a related self-compatible population (having the Sf gene). Plants carrying an SFI allele, by rejecting the Sf pollen, will restrict the spread of inbreeding and so be favoured by selection.

The origin of the SFI and Sf alleles are discussed in relation to the author's hypothesis of S-gene structure.

On the basis of chromosome behaviour in three strains of Drosophila melanogaster males—a wild-type strain, a translocation heterozygote strain and a double translocation strain—it is concluded that the cytologically detected apparent chiasmata in the autosomes are only surface associations of homologous chromosomes and do not involve exchange of chromosome material. Such chiasma-like associations occur also between non-homologous chromosomes. The stainable, Feulgen-positive material of the chromosomes is thought to be responsible for the surface associations.