Search

To examine whether structural and functional differences exist in the proliferation disrupter (prod) genes between Drosophila simulans and D. melanogaster, we analyzed and compared both genes. The exon-intron structure of the prod genes was found to be the same--three exons were interrupted by two introns, although a previous report suggested that only one intron existed in D. melanogaster. The prod genes of D. simulans and D. melanogaster both turn out to encode 346 amino acids, not 301 as previously reported for D. melanogaster. The numbers of nucleotide substitutions in the prod genes was 0.0747 +/- 0.0180 per synonymous site and 0.0116 +/- 0.0039 per replacement site, both comparable to those previously known for homologous genes between D. simulans and D. melanogaster. Genetic analysis demonstrated that D. simulans PROD can compensate for a deficiency of D. melanogaster PROD in hybrids. The PRODs of D. simulans and D. melanogaster presumably share the same function and a conserved working mechanism. The prod gene showed no significant interaction with the lethality of the male hybrid between these species.

The concepts of demography provide a means of combining the ecological approach to population growth with the genetical approach to natural selection. We have utilized the demographic theory of natural selection developed by Norton and Charlesworth to analyze life history schedules of births and deaths for populations of genotypes in Drosophila pseudoobscura . Our populations illustrate a stable genetic equilibrium, an unstable genetic equilibrium, and a case of no equilibrium. We have estimated population growth rates and Darwinian fitnesses of the genotypes and have explored the role of population growth in determining natural selection. The age-specific rates of births and deaths provide insights into components of selection. Both viability and fertility are important components in our populations.

Hybrids not carrying Drosophila simulans X chromosome derived from the cross between D. melanogaster females and D. simulans males are lethal at the late third instar larval period or early pupal stage. This lethality can be rescued by the mutation Lethal hybrid rescue (Lhr) of D. simulans. The Lhr gene is considered to play an important role in reproductive isolation, but the genetic analyses have not been carried out extensively because of the lack of visible mutations and chromosome rearrangements in D. simulans. Using a cuticle mutation, jabara, of which locus is close to Lhr, and D. melanogaster deficiencies, we performed cytological mapping in hybrids and estimated the Lhr locus to 54E-F.

Drosophila melanogaster populations in nature usually carry inversion polymorphisms. When they were transferred to and maintained in the laboratory as large cage populations, fre- quencies of polymorphic inversions were drastically decreased and finally eliminated. This "cage effect" was observed irrespective of the geographical origin of the population or the initial frequency of each inversion. The decrease and elimination of inversions in the cage was not overcome by changing conditions such as medium, temperature, or the number of isofemale lines (40-600) introduced. On the other hand, in the sets of isofemale lines derived from the same geographical origins as the cage populations, each of which was maintained as a small vial population, the inversion frequencies, though decreased from the initial frequencies, were kept at significantly high levels. The cage populations initiated with one or two isofemale lines also maintained the inversion polymorphisms that were as high as vial populations.

The eyeless Nanjo is a mutation that was found in a natural population of Drosophila melanogaster. Genetic analysis of this mutant revealed the following characteristics of this mutation. (1) The homozygous eyelessness is caused by a single recessive mutation which prevents the normal development of compound eyes. (2) An extremely variable morphological eye phenotype was generated by the introduction of a different genetic background to this mutant. (3) The estimated germ line reversion rate was very high. These results suggest that an extremely variable morphological eye phenotype might result from spontaneous somatic excision of some kind of a transposable element.

Interspecific crossability between two sibling species, Drosophila melanogaster and D. simulans, were examined in nature and in the laboratory. About 4.6% of D. melanogaster females were found to be inseminated by D. simulans males in six Japanese natural populations. In the laboratory experiments D. melanogaster females collected and established from non-sympatric or newly sympatric sites with D. simulans were more easily inseminated than those from historically sympatric sites. It may be suggested that the premating isolation does not develop rapidly after the sibling species come sympatry.

Changes in the genetic structure of the Katsunuma natural population of Drosophila melanogaster have been examined during the past 35 years. The frequency of recessive lethal genes on the second chromosome once increased from 15% to 30% in the early 1970s, then decreased to about 24% in the late 1970s, and thereafter showed no significant changes. Sterility genes, the frequency of which is always less than the lethals, showed a similar tendency. The SD (segregation distorter) mutant gene disappeared but some others such as rbl (reduced bristle) and bw (brown) persisted in the population. The frequency of inversion-carrying chromosomes gradually decreased in the period, such that the standard chromosome frequency in the second and third chromosomes increased from about 40% to more than 80%. Coincident with these frequency changes is the invasion of a transposable element P into the Katsunuma population. The P element should have invaded into Katsunuma in the late 1960s. It spread over the population apparently inducing deleterious mutations, causing the decrease in the allelism rate, and hence increasing the effective population size. Soon, however, most flies became resistant to the P element-mediated transposition as they began to harbor defective P elements. During the course of spreading, the P element must also have induced deleterious mutations on the polymorphic inversions, breaking up the heterotic gene complexes along the chromosomes, which probably caused the reduction in the frequency of inversion chromosomes. Temporal invasion of D. simulans, a sibling species of D. melanogaster, into Katsunuma occurred several times after 1978, and the species seems to have been settled since 1990. This, however, did not have any effect on the genetic structure of D. melanogaster population.

Hybrid females from Drosophila simulans females x Drosophila melanogaster males die as embryos while hybrid males from the reciprocal cross die as late larvae. The other two classes are sterile adults. Letting C, X, and Y designate egg cytoplasm, X, and Y chromosomes, respectively, and subscripts m and s stand for melanogaster and simulans, CmXmYs males are lethal in the larval stage and are rescued by the previously reported genes, Lhr (Lethal hybrid rescue) in simulans or Hmr (Hybrid male rescue) in melanogaster. We report here another rescue gene located on the second chromosome of simulans, mhr (maternal hybrid rescue) that, when present in the mother, rescues CsXmXs females from embryonic lethality. It has been postulated that the hybrids not carrying the Xs like CmXmYs males are larval lethal and that the hybrids carrying both the Cs and the Xm like CsXmXs females are embryonic lethal. According to these postulates CsXmYs males (obtained by mating attached-X simulans females to melanogaster males) should be doubly lethal, at both embryo and larval stages. When both rescuing genes are present, Hmr in the father and mhr in the mother, males of this genotype are fully viable, as predicted.

Lethal phases of the hybrids between Drosophila melanogaster and its sibling species, D. simulans are classified into three types: (1) embryonic lethality in hybrids carrying D. simulans cytoplasm and D. melanogaster X chromosome, (2) larval lethality in hybrids not carrying D. simulans X, and (3) temperature-sensitive pupal lethality in hybrids carrying D. simulans X. The same lethal phases are also observed when either of the two other sibling species, D. mauritiana or D. sechellia, is employed for hybridization with D. melanogaster. Here, we describe genetic analyses of each hybrid lethality, and demonstrate that these three types of lethality are independent phenomena. We then propose two models to interpret the mechanisms of each hybrid lethality. The first model is a modification of the conventional X/autosome imbalance hypothesis assuming a lethal gene and a suppressor gene are involved in the larval lethality, while the second model is for embryonic lethality assuming an interaction between a maternal-effect lethal gene and a suppressor gene.

Previous genetic studies have shown that wild-type function of the l(1)ogre (lethal (1) optic ganglion reduced) locus is essential for the generation and/or maintenance of the postembryonic neuroblasts including those from which the optic lobe is descended. In the present study molecular isolation and characterization of the l(1)ogre locus was carried out to study the structure and expression of this gene in order to gain information about the nature of l(1)ogre function and its relevance to the development of the central nervous system. About 70 kilobases (kb) of genomic DNA were isolated that spanned the region where l(1)ogre was known to reside. Southern analysis of a l(1)ogre mutation and subsequent P element-mediated DNA transformation mapped the l(1)ogre+ function within a genomic fragment of 12.5 kb. Northern analyses showed that a 2.9-kb message transcribed from this 12.5-kb region represented l(1)ogre. A 2.15-kb portion of a corresponding cDNA clone was sequenced. An open reading frame (ORF) of 1,086 base paris was found, and a protein sequence of 362 amino acids with one highly hydrophobic segment was deduced from conceptual translation of this ORF.

Eight species belonging to the Drosophila melanogaster subgroup were examined genetically and biochemically for the construction of phylogenetic trees. Based on the hypothesis advanced by Watanabe and Kawanishi (1979, 1981), we have estimated the ages for the eight Drosophila species in the descending order to be as follows: D, melanogaster, D, orena, D. simulans, D, yakuba, D, erecta, D. teissieri, D, mauritiana, and D. sechellia The plausible immediate ancestor for D. orena, D, simulans and D. sechellia appears to be D. melanogaster and that for D. yakuba, D, teissieri and D. mauritiana is D, simulans. On the other hand, a two-dimensional electrophoretic analysis has suggested that the subgroup consists of two complexes; they are D, melanogaster complex (D, melanogaster, D, simulans, D, mauritiana, D. sechellia) and D, yakuba complex (D, yakuba-D, teissieri pair and D. erecta-D. orena pair). Hybridization tests among the eight species have revealed that the most closely related pairs of species are the D, simulans-D, mauritiana within the complexes and the D, mauritiana-D, teissieri between the complexes.

Males of Drosophila melanogaster bearing various X chromosome segmental aneuploidies were tested for sensitivity/resistance to the male killing action of the Sex Ratio Organism (SRO). It was found that there is no region on the X chromosome which makes, when duplicated, males resistant to the SRO. A region which makes hetero-deficient females sensitive to the SRO also was not found. Searches for a gene (s) that can give males resistance to the SRO action were made using the X chromosomes continuously mutagenized for seven successive generations, the second and third chromosomes obtained from a natural population (100 lines), and the cytoplasm of 100 isofemale lines and twenty-two mass lines from various natural populations. No such genes were recovered. A possible reason for the failure of recovering such a gene is discussed.

1. The NSR-spirochetes injected to Drosophila simulans killed sons in the embryonic stage and the SR condition was established in this species by transmission of the spirochetes from mother to daughter through generations.2. Inter-species hybrid males between D. melanogaster and D. simulans were killed by the NSR agent mostly at the stages of larvae and pupae, while the hybrid females were not affected.3. Daughterless gene (da) killed the hybrid females in the embryonic stage.

Both cadmium and furylfuramide were highly toxic at every develompental stage of Drosophila, resulting in prolongation of the developmental time, lowering of the viability and decreasing of the egg production. However, these chemicals give no mutagenic effects either in the sex-linked or in the second chromosome lethal test.

1.The genetic variation of the hybrid crossability between Drosophila melanogaster females and D. simulans males was examined by using eight geographical strains of D. simulans from Japan.2. The hybridization was measured by insemination rate, egg to adult viability, and the number of hybrid progeny. The most experiments were undertaken with Oregon-R melanogaster females and various simulans male strains. The variation among simulans strains was clearly detected in these measurements and they were in large part positively correlated with each other.3.The temperature sensitivity of the hybrid varied from strain to strain of simulans, and was not ascribed to the geographical origin of the strain.4. D. melanogaster females were also variable in the hybrid crossability. Six melanogaster strains were significantly different in the insemination rate. A significant variance due to the interaction between melanogaster×simulans was also detected.5. The yellow mutant of melanogaster females showed an excellent receptivity to simulans males, and the yellow locus itself seemed to respond it.

Through examination of all available data on lethal and inversion frequencies on the second chromosome in natural populations of Drosophila melanogaster, we have discovered that there is a clear negative correlation between the two quantities. Lethal genes are located more densely on the regions of standard gene arrangement than the inverted regions, and this accounts for the negative correlation. To reveal the underlying mechanism of the phenomena, we have carried out an experiment and found that effect of EMS-induced mutations on the inversion-carrying chromosome is more severe than that on the standard chromosome. We interpret these results as evidence for coadaptation or position-effect within the inversion chromosomes. New mutations within the coadapted gene complex are quickly eliminated from the population and polymorphic inversions are kept free of mutants through selective elimination.

A mutant of a strain of the SR organisms (SRO, transovarially transmitted male-killing spiroplasmas) of Drosophila is described. The mutant, NSRO-A, has lost the male-specific killing activity on host flies, but maintains other properties such as the clump forming spectrum and the extractable titer of the associated virus as original NSRO strain.

SummaryWithin-line heterogeneity has been found in the mitochondrial DNA (mtDNA) in two isofemale lines ofD. simulans. The co-existing types, S and M, were typical of the mtDNA inD. simulansand inD. mauritiana, respectively, their nucleotide divergence per site beingca.2·1%. Segregation analysis confirmed that some individuals in these lines were heteroplasmic and suggested incomplete maternal inheritance of mtDNA inDrosophila. Examination of other lines ofD. simulansrevealed that the M type ofD. mauritianaoccurs at 71% in Réunion, 38% in Madagascar and 0% in Kenya. This finding and interspecific sequence comparisons of both M types indicate thatD. mauritianadiverged fromD. simulansprobably less than 240000 years ago.

Intra-specific variations of the courtship song (inter-pulse interval: ipi) of Drosophila melanogaster and D. simulans males were investigated. The variation among strains was small in D. melanogaster (31.3-34.8 msec), whereas it was much larger in D. simulans (42.8-56.8msec). The average ipi was 32.8msec for D. melanogaster and 49.1msec for D. simulans, and the distributions between the two sibling species were non-overlapping. However, the ipi itself plays less important role for the success of mating.

Two different species which were originated from India and Indochina have been called Drosophila jambulina as the same species. They were compared genetically, cytologically and electrophoretically, as well as morphologically. From these results it was concluded that the Indian species was the realD. jambulinaand that the Indochinese species seemed to be a new species and was tentatively namedD. punjabiensis-likebecause it was closer to the sibling species, D. punjabiensis. D. jambulinashowed a color dimorphism in a terminal tergite of female. A light-type gene controlling the dimorphism was dominant over a dark-type gene. The heterozygote segregated into 3 light: 1 dark in the next generation.

Selection for geotaxis was carried out with flies from a natural population of Drosophila melanogaster; geotactic behavior was measured by means of a Hirsch classification maze. The population was initally almost neutral to gravity, and it responded to both positive (downward) and negative (upward) selection with a realized heritability of about 0.13. Stabilizing selection toward neutral gravity was carried out simultaneously. At generations 6, 9, and 10, all possibly hybrid crosses between pairs of the selected populations were generated and tested. The geotactic scores of hybrids in generations 6 and 9 were not significantly different from the midparent values, while the scores of hybrids in generation 10 deviated significantly from the midparent values in the direction of positive geotaxis. The frequencies of polymorphic inversions declined in every population during selection, but the population under neutral selection seemed to maintain a higher chromosomal polymorphism than those under positive or negative selection. There was no significant depression of productivity, measured as number of progeny, in any population during nine generations of selection.

Approximately 100 abundant proteins were analyzed by two-dimensional electrophoresis (2DE) in eighteen species belonging to three taxonomically differentDrosophila groups, theD. melanogaster species subgroup, theD. auraria complex and theD. virilis species group. Genetic distances between these species were calculated from a total number of protein differences in the pairs of species. Based on such data of genetic distances a dendrogram was constructed. The genetic distances or phylogeny obtained by 2DE were significantly correlated with those for one-dimensional (conventional) electrophoresis and also with data on mating success between species. It was discussed on the basis of these studies that a technique of 2DE is a useful adjunctive tool in systematics, including the previous reports with different kinds of data.

Further characterization of Lhr gene, which rescues the lethal hybrid of D. melanogaster and D. simulans, was made using several mutants of D. melanogaster. It was known that hybrids carrying the Y chromosome of D. simulans are lethal without exception, that is, hybrids carrying X chromosome of D. simulans are all viable except a female hybrid of female simulans and male melanogaster. Lhr rescues all the lethal hybrids carrying simulans Y (=hybrids without carrying simulans X). Brief discussion on the character of hybrid lethality between the two species and how the Lhr works are presented.

The frequencies of a polymorphic inversion, In(2L)t, and of Adh and αGpdh alleles were analyzed in three natural populations of Drosophila melanogaster from Japan. Significant positive correlations between the frequencies of In(2L)t and AdhS or ?GpdhF were detected due to tight linkage. An analysis of correlation with latitude showed that the negative cline of AdhS frequency could be explained entirely by its linkage with In(2L)t; the frequency of AdhS on the standard chromosome did not show a latitudinal cline. To the contrary, the cline of αGpdhF frequency itself was positive, and its linkage with In(2L)t makes the positive cline unclear. These results suggest that the two allozymes themselves respond to latitudinal natural selection in different ways. When these populations were transferred to laboratory cages and maintained for a long time, they lost the chromosomal polymorphism but retained stable enzyme polymorphisms, although allele frequencies in the cage were not the same as in nature. The frequencies of Adh and αGpdh alleles were close to those in earlier cage populations of the same geographical origin.

A gene, Lethal hybrid rescue (Lhr 2-95), which rescues the lethal hybrids between Drosophila melanogaster and D. simulans was found on a second chromosome of D. simulans from a natural population. The cross, melanogaster _??_×simulans _??_, which normally produces only female progeny, resulted in the production of both female and males at an equal frequency when the simulans males carried Lhr. The reciprocal cross, simulans _??_×melanogaster _??_, which ordinarily produces only males, gave rise to about 14% females and 86% males. The cross, XX/Y melanogaster _??_×simulans _??_ carrying Lhr, produced a few percent females, although normal simulans produced only hybrid males. The sex ratio (SR) factor and daughterless (da) gene killed either hybrid males or hybrid females according to their usual actions, showing no specific interactions with the Lhr gene. All hybrids, those rescued by Lhr as well as those produced normally, were completely sterile. Evolutionally implications of Lhr are briefly discussed.

Photo-preferences of the sibling species, Drosophila simulans and D. melanogaster, were studied using an apparatus and population cage having a gradient of light intensities. D. simulans preferred to stay and to lay eggs at light places whereas D. melanogaster did not show such a strong light preference. Selections of eggs in the light gradient cages made it possible to segregate a mixed species population into different species populations; selection for photo-positive flies soon eliminated D. melanogaster and slection for photo-negative or photo-neutral flies finally eliminated D. simulans. The difference of the photo-preference seems to be a cause of coexistence of the sibling species in nature.

Drosophila simulans has been found for more than 10 years in Japan, mainly in suburban areas whereas its sibling species D, melanogaster tends to be found in houses. Seasonal activities of the two species were clearly different; In a typical hot summer the frequency of D. simulans decreased temporarily in August making two peaks in July and September. But in a cool summer the two peaks disappeared making a single peak in September or October. On the other hand, D. melanogaster showed higher activity in August in the hot summer but it occasionally shifted the active peak during the season avoiding competition with D. simulans. A nine year survey of D. simulans around the area of Mt. Fuji revealed that the species gradually invaded from the Pacific coast to the inland. Population size in autumn was strongly affected by the temperature of the Preceeding winter. Cold winters such as in 1977 and 1981 decreased the size drastically at every collection site.

The P-M system of hybrid dysgenesis was investigated in two Japanese regions, Katsunuma and Ohzu. Most strains were Q in that they did not produce appreciable gonadal dysgenesis with P males or with M females. A few strains produced ambiguous results. Some of these were unstable and on later testing were P, Q or M. Eleven laboratory strains which had been derived from various sites in Japan and had been maintained for 8 to 27 years were also examined. Eight (including all four that were 27 years old) were M, two were Q, and one which had been kept as a large cage population showed some P activity.

Phylogenic relationships among 21 members (15 species) of the Drosophila montium species subgroup were investigated by O'Farrell's two-dimensional electrophoresis (2DE) and by starch gel electrophoresis (SDE). Phylogenic positions of two unclassified species [D. punjabiensis-like and D. sp. (Iriomote)] as well as those of two new species (D. bocki and D. leontia) within the D. kikkawai complex could be assigned, and it was evidenced that both electrophoretic techniques are adjunctively useful tools in taxonomy. Although the magnitude of genetic differentiation may be different between protein loci surveyed by 2DE and allozyme loci obtained by SDE, the relationship of genetic distances was highly correlated between two data sets of 2DE and SDE, as shown previously by Aquadro and Avise (1981). The biochemical dendrograms in the subgroup were constructed by the data sets (2DE and SDE) of genetic distances between species. The genetic relationships were consistent with the results of interspecific hybridization among these species.

A recessive, non-maternal effect, female-specific lethal mutant of Drosophila melanogaster was recovered from a natural population. The mutant, fle(3)100, is located at 3-45±. About one-half of homozygous females show normal development up to the pharate adult stage. Salivary gland chromosomes of third instar homozygous female larvae have normal staining properties in the lactic-aceto-orcein staining, suggesting normal dosage compensation. The results indicate that the fle(3)100 gene forms a new class of sex-specific lethals.

SUMMARYThe number of loci which are potentially able to produce sterility genes was estimated forDrosophila 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.

Fertility, the component of selection due to female fecundity and male mating success, differed significantly among the ST/ST, ST/AR, and AR/AR karyotypes in experimental populations and varied with karyotypic frequency. In relation to ST/AR, ST/ST females and males had higher fertilities at low frequency; AR/AR males and females were at a significant fertility disadvantage at intermediate frequency, while at low and at high frequencies their fertilities matched or exceeded that of the heterokaryotype. These fertility differences were comparable in size to viability differences previously reported for D. pseudoobscura karyotypes. Differential fertility seems likely to be an important element, perhaps just as important as differential viability, in the balancing selection that maintains the chromosomal polymorphism in this species.

Four common cosmopolitan inversions (2Lt, 2RNS, 3LP, 3RP), four rare cosmopolitan inversions (2LA, 3LM, 3RC, 3RMo) and two recurrent endemic inversions (2LW, 3LY) were found to be distributed in Japanese natural populations of Drosophila melanogaster. Moreover, seventeen unique endemic inversions were rather randomly detected in many natural populations as well as in stocks that originated from small number of wild caught flies. The frequency of inversions was generally higher in Southern populations than in Northern populations, especially in the case of In(3R)P. The polymorphic inversions did not show any heterosis in the present analysis. The frequencies of inversion heterozygotes were almost equal to the expectaions with a few exceptions of a scarcity of the heterozygotes over the expectation. In the Katsunuma population, some inversions decreased remarkably within ten years, which resulted in the change of abundance order among four common cosmopolitan inversions. This kind of change might have occurred in the Shiojiri population. Both populations are very interesting in that new inversions are increasing in frequency and replacing the old ones.

Seventeen species of the Drosophila montium subgroup, originated from the Southeast Asia, were genetically examined to clarify the phylogenetic relationships. Among 272 interspecific crosses, 61 combinations were successful in mating and 39 combinations produced hybrid flies. These results enabled us to classify the subgroup into three species complexes: the kikkawai complex (6 species), the jambulina complex (4 species) and the auraria complex (7 species), which were very similar to that obtained by the electrophoretic classification (Ohnishi and Watanabe, 1984). Asymmetrical mating preference between species was found in the present experiment. They were applied to estimate the relative age of species according to the hypothesis proposed by Watanabe and Kawanishi (1979). The evolving order was as follows: kikkawai, leontia, pennae, lini-like, lini, bocki in the kikkawai complex, punjabiensis, punjabiensis-like, jambulina, barbarae in the jambulina complex, and quadraria, yuwanensis, rufa, subauraria, biauraria, triauraria, auraria in the auraria complex. Hybrid flies, if produced, were mostly fertile or partially fertile in both sexes (32/39) within species complex crosses. Therefore premating isolation played a more important role than postmating isolation in speciation of the D. montium subgroup.

Four Drosophila species, D. baimaii, D. bicornuta, D, burlai and D, truncata, belonging to the montium species subgroup, were examined for the abdominal tergite coloration. Females show dimorphsim, light or dark, in coloration. In all species examined, the color dimorphism is regulated by two alleles of a single autosomal locus and the dark allele is dominant. Males of the first three species show monomorphic tergite coloration, either light or dark, while males of the fourth species, D. truncata, show the light/dark color dimorphism. Here again it is controlled by two alleles of a single autosomal gene and the dark allele is dominant.

Chromosomal polymorphism in natural populations is largely maintained by the adaptive superiority of inversion heterozygotes. Dobzhansky and his co-workers have carried out many studies on heterosis in Drosophila pseudoobscura populations (summarized in Dobzhansky, 1970). The superior fitness of heterokaryotypes has also been reported in other Drosophila species: D. persimilis (Spiess, 1958; Spiess and Langer, 1964); D. robusta (Carson, 1958; Prakash, 1967); D. willistoni (da Cunha et al., 1950; Pavan et al., 1957); D. ananassae (Tobari, 1962) and D. pavani (Brncic et al., 1969). Extensive chromosomal polymorphism in Japanese populations of D. melanogaster was demonstrated by Watanabe (1967). Two inversions on the second chromosome and four inversions on the third chromosome which had been found all over the world were highly polymorphic in Japanese natural populations. The second chromosome inversion In(2L)B was maintained in the population year after year at a frequency of about 35%. A higher adaptive value of the inversion heterozygotes was expected from estimates of the pre-adult viability. However, superiority was found only when In(2L)B was combined with another inversion, In(2R)C. The viability of the heterozygotes for In(2L)B and standard chromosomes was the same as that of the standard heterozygotes (Watanabe et al., 1964). A significant superiority of this inversion heterozygote in female

Three or four percent of the wild flies in natural populations of D. melanogaster have been found to be sterile. An analysis of sterility associated with the second chromosome revealed a much lower frequency of genetically sterile flies. The accumulation of sterility genes in a cage population was proportional to that of lethal genes, as were their equilibrium frequencies in several natural populations. Many sterile chromosomes were associated with low viability due to pleiotropic effects. The number of chromosomes leading to sterility in both sexes was larger than the expectation based on random combination of male and female sterility genes. This suggests that there is some linkage disequilibrium between male and female sterility genes, as well as a pleiotropic effect of single sterility genes. Some sterility genes were maintained in natural and cage populations, and the patterns of persistence of the sterility genes were very similar to those of lethal genes.

Genetic variabilities in Japanese natural populations of D. melanogaster distributed at Kofu-Katsunuma locality have been analyzed. The frequency of lethal and semilethal chromosomes was 21.5 per cent in coadapted genetic background and 36 per cent in Samarkand isogenic background during the past seven years. The effect of the genetic background on the viability of homozygotes was remarkable for homozygous semilethal and subvital chromosomes. This means that a coadaptive interaction may exist between chromosomes.A clear negative correlation between the frequency of deletelious chromosomes and ‘relative’ viability of quasinormal homozygotes was observed in Drosophila populations including Kofu-Katsunuma.Allelic rates between lethal genes extracted from the natural populations fluctuated during the past several years. The average allelic rate was 3.48 par cent. The low frequency of deleterious genes and high frequency of allelism was assumed to be characteristic for Far East and Russian populations of D. melanogaster.The frequency of elimination (IQ2) caused by deleterious genes in homozygotes was estimated to be 0.0016 in average Kofu-Katsunuma population which was comparatively higher than in other natural populations. On the other hand, the mutation rate (UL+SL) of lethal and semilethal genes on the second chromosome was assumed to be 0.0075, so that the fitness of heterozygotes for deleterious genes should be, on the average, inferior to that of normal heterozygotes and they would be selected out from the natural populations in Japan.

1. Segregation distorter genes (SD) were found in Kofu and Katsunuma natural populations of Drosophila melanogaster.2. The nature of SD gene extracted from Kofu and Katsunuma populations was almost the same as in Madison; segregation distortion occurred only in SD heterozygous males and it did not occur either in SD homozygous males or in females.3. SD gene was mostly found in association with a right arm inversion, In(2R)C, and some SD chromosomes simultaneously carried the same persistent lethal gene which was frequently found and maintained in Kofu and Katsunuma natural populations.4. The distance between the persistent lethal gene and SD gene is so short that its retention in the population may be effectively due to its association with SD. Heterotic gene complexes including the lethal and SD genes were discussed as a persistent mechanism.5. The sensitivity of the second chromosomes to SD action varied from complete sensitiveness (k=1.0) to resistance (k=0.5). The mean k value of many chromosomes isolated from Kofu and Katsunuma populations was 0.77.

Differences in the degree of successful matings between pairs of reciprocal hybrid matings indicate the direction of evolution among related species of Drosophila. Females of a derived species do not mate with males of the ancestral species, whereas females of the ancestral species readily mate with males of the derived species.

Genetic changes with time are frequently observed in natural populations of Drosophila. Dobzhansky (1971) demonstrated two types of genetic changes in frequencies of polymorphic third chromosomes in D. pseudoobscura and D. persimilis. One is the seasonal cyclic changes and the other is changes over several years. The former was largely due to natural selection with the populations responding to seasonal changes in their environments. The latter was more or less directional and the causation was rather vague. Dobzhansky speculated that the changes were induced by pesticides such as DDT in the environment or by the emergence of new supergenes in the chromosome inversions. Seasonal or annual changes in gene frequencies, especially for lethal genes, have been reported by many investigators who censused natural populations of D. melanogaster. Dubinin (1946) showed in Russian populations an increase of lethal frequency during each breeding season. Such changes were later referred to as the "Dubinin cycle" and have been confirmed in a Japanese population (Minamori and Saito, 1964). However, most seasonal surveys failed to show a clear "Dubinin cycle" (Ives, 1954; Goldschmidt et al., 1955; Kosuda et al., 1969). The temporal shifts of lethal genes in the Massachusetts natural population for the past 30 years have been reported by Ives (1945, 1954, 1970), Band and Ives (1961, 1963, 1968) and Band (1969, 1972a, 1972b). These authors found a correlation of the frcaniuncv chanP's with certain weather factors; positively with summer rainfall and negatively with the average daily temperature range of the week prior to collection. Hoenigsberg et al. (1969, 1973) observed a niche-dependent change of lethal genes in the American tropics. The periodic use of pesticides was considered as a major cause. Recently Minamori et al. (1973) reported a sudden increase of lethal genes in the Hiroshima, Japan, population. The effects of weather factors proposed by Band (1972a) were not evident. Their interpretation is that the environment of Drosophila became favorable to its overwintering. Genetic variation in the Katsunuma, Japan, population in the 1959-1968 period has been reported by Watanabe (1969b). The frequency of lethal second chromosomes was about 15% in that period. Two polymorphic inversions, In(2L)B and In (2R)C, had been maintained in the population at high levels of about 33% and 25% during the early sixties (T. Watanabe, 1967). The purpose of this paper is to report an increase of lethals accompanied by a decrease of the inversions in the population after 1969.

A simultaneous survey of 14 protein loci, together with frequencies and within- and between-population allelism rates of lethal chromosomes, was carried out in five (four Japanese and one Korean) natural populations and one cage population of Drosophila melanogaster. It was found that lethal allelism rates decrease rapidly as geographic distance between two populations increases, while variation at protein loci shows a remarkable similarity over all populations examined. These findings suggest that there are very high levels of gene flow in these natural populations and that selection at protein loci which can maintain substantial geographic variation, if present, is overshadowed by gene flow. There is no indication that invasion of D. melanogaster to the Far East occurred so recently that the frequencies of lethal chromosomes are still in nonequilibrium.

Five hundred and two third chromosomes were extracted from a large cage population of Drosophila melanogaster initiated two months after collection of the progenitors near Raleigh, North Carolina in 1970.—Salivary gland chromosomes of 489 chromosome lines were examined and 54 chromosomes were found to carry inversions. The inversions were classified into three polymorphic types [In(3L)P, In(3R)P, and In(3R)C] and two unique types. The polymorphic inversions were found in frequencies of 0.012, 0.088, and 0.010, respectively.—Viabilities of homozygotes and heterozygotes were examined. Chromosomes with lethals occurred with a frequency of 0.495: 0.537 in the group of inversion-carrying chromosomes and 0.490 in the group of inversion-free chromosomes. The average homozygote viability computed on the basis of an average heterozygote viability of 1.0000 was 0.3235 if lethal lines were included and 0.6290 if they were excluded. The detrimental load to lethal load ratio (D:L ratio) was 0.70 (=0.4636-0.6650). The average viability of lethal heterozygotes was significantly larger than that of lethal-free heterozygotes. It appears, however, that lethal genes in heterozygotes have deleterious effects on fitness as a whole.—The average degree of dominance for viability polygenes was estimated to be about 0.3-0.4 in lethal-free individuals and nearly zero in lethal heterozygotes. Overdominance or some form of balancing selection was suggested at some loci. The difference between the values obtained for average degree of dominance due to genetic backgrounds and superior vibaility of lethal heterozygotes (but not fitness as a whole) suggests that some epistasis or coadaptation occurs.—The results described above are similar to those obtained for the second chromosomes (Mukai and Yamaguchi 1974).

Two hundred and ninety second chromosomes extracted from a natural population of Drosophila melanogaster were analyzed to estimate the genetic variance of viability and its components by means of a partial diallel cross (Design II of Comstock and Robinson 1952). The additive and dominance variances are estimated to be 0.009 and 0.0012. Using the dominance variance and the inbreeding depression, the effective number of overdominant loci contributing to the variance in viability is estimated to be very small, a dozen or less. Either the actual number of loci is small, or the distribution of viabilities is strongly skewed with a large majority of very weakly selected loci. The additive variance in viability appears to be too large to be accounted for by recurrent harmful mutants or by overdominant loci at equilibrium with various genetic parameters estimated independently. The excess might be due to frequency-dependent selection, to negative correlations between viability and fertility, or possibly to the presence of a mutator. The selection for viability and fertility, or possibly to the presence of a mutator. The selection for viability at the average polymorphic locus must be very slight, of the order of 10-3 or less.

Drosophila melanogaster is a chromosomally polymorphic species, a fact that has been confirmed in various natural populations by many workers (Dubinin et al., 1937; Warters, 1944; Ives, 1947; Carson, 1965; Yang and Kojima, 1972; Mukai and Yamaguchi, 1974; Ashburner and Lemeunier, 1976; Stalker, 1976; Choi, 1977; Mettleretal., 1977; Inoue and Watanabe, 1979; Paik, 1979; Zacharopoulou and Pelecanos, 1980; Knibb et al., 1981; Knibb, 1982). Chromosome variations in nature are usually restricted to paracentric inversions on the four arms of the two major autosomes. Other variations such as X-chromosome and pericentric inversions, deficiencies, duplications, translocations and transpositions rarely are found in nature. Inoue and Watanabe (1979) classified the inversions into four groups, using a classification partly modified from that of Mettler et al. (1977): (1) Common cosmopolitan inversions which are the most frequent world-wide inversions on each arm of the major autosomes. (2) Rare cosmopolitan inversions which are distributed world wide but in lower frequency than the common cosmopolitan inversions. Some populations often lack one or two of these inversions. (3) Recurrent endemic inversions which show temporal polymorphisms in some populations. (4) Unique endemic inversions which are observed in a single individual or its brood from a single population and never found in other populations. Designations of the common cosmopolitan inversions are In(2L)t, In(2R)NS, In (3L)P and In(3R)P. The rare cosmopolitan inversions are labelled In(2L)A, In(3L)M, In(3R)C, In(3R)Mo and In(3R)K. In Japan two inversions, In(2L) W and In(3L) Y, are categorized as recurrent endemic inversions. The unique endemic inversions are often reported as rare inversions, having differing break points which probably correspond to recent mutations. Mettler et al. (1977), in their studies of eastern United States populations, demonstrated highly significant negative correlations between the frequencies of the common cosmopolitan inversions and latitude. Stalker (1980) obtained similar results in addition to documenting seasonal changes of some inversions in midwestern and eastern U.S. populations. Recently, Knibb et al. (1981) reported a north-south cline from Australian populations with inversion frequencies increasing toward the equator. These studies suggest that the standard arrangementbearing chromosomes were more adaptive in cooler environments than were the inversion-containing chromosomes. Inoue and Watanabe (1979) compared the inversion frequencies of several Japanese populations showing a north-south dline in inversion In(3R)P. Clines were not always confirmed in other inversions which was attributed to the small number of populations examined. Inoue and Watanabe (1979) also compared several natural populations from the viewpoint of frequency order of the four common cosmopolitan inversions: In order of decreasing frequency the relation 2Lt > 2RNS > 3RP > 3LP was observed in the middle part of Japan in the past (1960's) and still (1970's) in the northern part of Japan. However, the present populations in the middle part of Japan showed the relation of 3RP > 2Lt _ 2RNS > 3LP. The order change in the inversion frequencies occurred

Lethal and sterility mutations were accumulated in a cage population which was initiated with lethal- and sterility-free second chromosomes of D. melanogaster. It took about 2,000 days for the frequencies of these genes to reach equilibrium levels, i.e., 18% lethal and 9% male-sterile chromosomes. Two other cage populations which were initiated with random chromosomes sampled from natural populations and kept for more than eleven years in the laboratory showed 19-20% lethal content. The elimination rates of lethals by homozygosis in these populations were smaller than the mutation rate. By using Nei's formulae, the deleterious effect of a lethal gene in heterozygous condition (h) was estimated to be 0.035. The effective population number in the cage populations was estimated to be 1,000-2,900, while the actual population number was 3,500-7,800.