Eighty-three strains of Setaria italica collected from various areas throughout Eurasia were crossed with three tester strains from Japan (tester A), Lan Hsu Island of Taiwan (B) and Belgium(C). The F₁ hybrids were obtained in 224 of the 249 cross combinations tried. The pollen fertility of the F₁ hybrids ranged widely from almost sterile (8.2%) to highly fertile (99.1%) while there was a little variation among the individual F₁ plants of the same cross combination. The 62 strains, which successfully produced F₁ hybrids with all of the three testers, could be clearly classified into six types, designated as types A, B, C, AC. BC and X. Pollen fertility of more than 75% was regarded as normal. The strains of types A, B and C were those which produced F₁ hybrids having normal pollen fertility when crossed with testers A, B and C, respectively. When both F₁ hybrids from the crosses with two testers, A and C, or B and C, showed normal pollen fertility, the strain was classified as type AC or BC. The strains whose F₁ hybrids always showed pollen fertility of less than 75% in all three cross combinations were designated as type X. There was no strain whose F₁ hybrids always showed normal pollen fertility in the cross combinations with all three testers or both testers A and B. These types classified are thought to reflect the landrace groups phylogenetically differentiated, because hybrid sterility occurs as the results of genetic differentiation between strains. Most type A strains were distributed in East Asia including Japan, Korea and China. Type B strains were narrowly found in Taiwan and in the southwestern part of the Nansei Islands of Japan. Most European strains were found to be type C. Strains of types AC and BC were distributed in Afghanistan and India, respectively. Types AC and BC are thought to be less specialized genetically than A, B or C. The geographical distribution of these landrace groups suggests that S.italica was first domesticated in an area ranging frorn Afghanistan to India, and then dispersed both eastward and westward from there.

View full abstract

New techniques using the carotene bleaching activity of. Iipoxygenase were developed for the detection of soybean seed lipoxygenases, L-1, L-2, and L-3 isozymes, respectively. The carotene bleaching tests for the three isozymes were performed as follows : (Test I for L-3) the seed extract was incubated with linoleic acid and carotene at pH 6.8. When the seed extract contained L-3, carotene was bleached in a few minutes at 20°C. In the absence of L-3 carotene bleaching was not observed ; (Test II far L-2) the seed extract was incubated with arachidonic acid and carotene at pH 6.8. After incubation for 3 minutes at 20°C, methanol was added into the mixture to stop the bleaching reaction. Since L-2 is much more active than L-1 and L-3 towards arachidonic acid at pH 6. 8, the lack of L-2 significantly decreased the level of the carotene bleaching activity. The presence or absence of L-2 in the F₂ seeds from Suzuyutaka (Lx2/Lx2) x P186023 (lx2/lx2) soybeans was accurately determined by the application of Test II ; (Test III for L-1) the seed extract was preincubated with linoleic acid and carotene at pH 9. 0. After preincubation for 60 seconds at 20°C, the mixture. was adjusted to pH 6.8. When the seed extract did not contain L-1, carotene bleaching was observed within thirty minutes at 20°C. The levels of the carotene bleaching activity in the three tests were determined visually or by measurements in a spectrophotometer at 452 nm within 60 minutes. Since the L-3 isozyme is indispensable for the carotene bleaching activity, seed extracts from soybeans la.cking L-3 do not cooxidize carotene. Therefore, in order to determine the presence or absence of L-2 or L-1 in the seeds lacking L-3, seed extracts from the soybeans lacking L-2 and L-1 in addition to the seed extracts to be examined were added to 60mpensate the lack of L-3 isozyme into the reaction mixture, respectively. The other procedures were the same as those applied in Tests II and III, respectively. These tests which are simple, rapid, and accurate, require only a few mg of seed sample.

View full abstract

This study was carried out to determine content of sugars, water soluble polysaccharides (WSP) and starch during the first 4 weeks after pollination of sugary. (su) and shrunken (shr-1s, shr-1a and shr-2) mutants of rice. The content of reducing sugars and sucrose in shr-1s, shr-1a and shr-2 mutants was high through early stages of endosperm development. Sugary mutant possessed higher WSP than the other mutants. Starch content in su, shr-1a and shr-2 reduced to about one-half of the parent cultivar. The starch reduction in shr-1s genotype was in-between the other mutant lines and the parent cultivar. These results indicated that the three loci affect carbohydrate production in the endosperm of rice.

View full abstract

Effect of hormonal balance in the culture media was investigated for getting reproducible method in protoplast culture of cabbage, Brassica oleracea L. Different hormonal combinations in the media were tested at each step of protoplast culture, i. e. cell division in liquid media, callus formation on agar media, and shoot regeneration from callus on agar media. Frequency of shoot regeneration was affected by phytohormones not only in the regeneration media (a final step), but also in the media for callus formation step. Callus developed in the medium with 0.5 mg/l 2, 4-D and 5 mg/l kinetin showed the highest shoot regeneration capacity, while those developed in the medium with 2 mg/l 2, 4-D and 1mg/l kinetin regenerated few shoots. Callus cultured with NAA as an auxin showed relatively low regeneration capacity compared that with 2, 4-D, and sometimes differentiated roots before transplanting onto the regeneration medium. Calluses with roots appearing first resulted in very poor shoot regeneration. From these results, shoot regeneration capacity of calluses derived from protoplasts was considered to be determined at an early stage of culture.

View full abstract

Using eu- and alloplasmic lines of a common wheat cultivar, Chinese Spring, and euplasmic line of cv. Norin 61, the effects of four fertile cytoplasms derived from Aegilops squarrosa, Ae. speltoides, Ae. kotschyi and Ae. vavilovii on the performance and interplant variability of 13 agronomic characters in the F₁ and F₂ generations of the cross between Chinese Spring and Norin 61 were investigated. The performances and interplant variabilities of both generations were little affected by any of the alien cytoplasms, but sterility increased slightly by the vavilcvii cytoplasm in the F₂ generation. Transgressive segregation attributable to the nucleus × cytoplasm interaction could not be detected in the present F₂, in contrast with some positive transgressive segregation observed in the F₂ of the cross between two related cultivars, Norin 61 and Norin 26, under the presence of the speltoides and kotschyi cytoplasms (TSUNEWAKI et al. 1985). Neither the heterosis nor the heritability in a broad sense was affected significantly by any alien cytoplasm. The present and previous results suggest that new common wheat genotypes better adapted to an alien cytoplasm than to the wheat cytoplasm might be created step by step, using the hybrids between closely rather than distantly related cultivars.

View full abstract

Occurrence of meiotic crossover and transmission of recombined chromosomes in selfed generations after a cross of two pure lines were simulated by applying the Monte Carlo method and the changes in the magnitude of heterozygosity and in the length of parental linkage blocks with the generations were examined. The number of generations required for the fixation of all the genes on a chromosome was proportional to the logarithm of the number of crossovers per chromosome (NCR). Increase in NCR from 1 to 10 resulted in the delay of fixation generation from 4.25 to 7.32. A cumulative distribution curve with NCR=z for the frequency of chromosomes homozygous for the whole length was roughly equivalent to the curve for independent factors with the number of factors equal to z multiplied by 3.3 to 4.0. The number of generations needed for the fixation of the whole genome depended upon the number of pairs of homologous chromo-somes involved in the genome as well as NCR. When NCR=2, for instance, the generation when 95% or more of the individuals come to complete fixation is Fg for a single pair of chromosomes, while F₁2 and F₁8 for a set of 7 and 12 pairs, respectively. Delay of fixation generation with increase of NCR was similar irrespective of the number of chromosomes. The average number of breakups of the linkage blocks increased from 0.95 to 7.31 in proportion to the increase in NCR from 1 to 10 and the maximum number ranged from 5 to 17. For a particular value of NCR, there was a significant positive correlation between the number of generations needed for fixation and the number of breakups. When 1% of the individuals with the highest number of breakups was selected, the average number of breakups was 2 (NCR=1) to 5 (NCR=10) times that of the original, while the number of generations needed for fixation changed into F9 or so irrespective of NCR. Increase in NCR only in a single generation was effective in raising the number of breakups when the increase occurred in early generations. It was found moreover by simulation that the fixation of the whole genome in as early as F₂ by applying the technique of anther culture and doubling of chromosome number resulted in a narrower distribution of the number of breakups and highly recombined genotypes could not be obtained.

View full abstract

The spatial pattern of variation in the leaf shape of the Japanese populations of the wild radish (Raphanus sativus L. var. raphanistroides MAKINO) was investigated by examining herbarium specimens, natural and cultivated samples. Three genetically determined leaf forms were included in natural populations of the wild radish. The pinnatifid (PF) leaf form was predominant, the lyrate (LY) leaf form was observed in a low frequency while the pinnatisect (PS) leaf form was rarely found. The progeny tests suggested that the LY leaf form is controlled by a single gene which is recessive to its alleles specifying the PF form. The proportlons of leaf morphs in the wild radish were similar to those in the Japanese radish, while the localization of the LY morph was detected in latitudinal and microgeographic scales. The frequency of the LY morph and magnitude of polymorphism increased with the increase in the latitude of the locality of the population. A high level of diversity for the leaf shape was recognized in many populations, while ecological preference was not detected among the different types.

View full abstract

To breed the sweet potato varieties adapted to true seed planting culture, open-pollinated population composed of natural flowering straines was made by the procedure as mentioned in the previous paper (Yamakawa and Sakamoto 1980), and recurrent selection for seed-setting and adaptability to true seed planting was conducted for two cycles. First, the yielding traits of polycrossed progenies of the parental straines with an adequate number of seeds under stick training culture were evaluated under true seed planting culture, and half-sib families with high total root weight and high dry matter percentage were selected. Next, the individual roots within selected half-sib families were selected for root size and root shape in order to make the open-pollinated population for the next cycle (Table 1). Population means of seed-setting tended to increase, but there were no significant differences between cycles. Total root weight decreased on the first cycle, but it increased over the basic cycle on the second cycle to 1538 kg/ha (182 kg/ha increase per cycle). Dry matter percentage decreased at the rate of 1.1% per cycle and reached 31.7% on the second cycle. The total root weight of the parental clone population under transplanting culture increased, but its dry matter percentage decreased as well as half-sib family population under true seed planting culture (Table 2). Genetic variability of all traits except germination percentage tended to decrease from the basic cycle to the second cycle. But genetic components of variance on the second cycle were still significant (Table 4).

View full abstract

For practical use of F₁ hybrid in Japonica rice it is necessary to increase natural out-crossing rate and seed set percentage on cytoplasmic male sterile lines in hybrid seed production field by reforming those floral characteristics which do not normally fit cross-pollination. Floral characteristics of some wild species and Indica cultivars of rice are estimated to fit natural outcrossing (OKA and MORISHIMA 1967, PARMAR et al. 1979a). Although introduction of these characteristics from wild rice or Indica to Japonica is one way of increasing natural outcrossing in Japonica rice, it is too difficult to introduce only these floral characteristics owing to the strong linkage between the floral characteristics and undesirable wild rice traits (IRRI 1985). In addition, Indica-Japonica hybrid shows hybrid sterility (MORINAGA 1968). Accbrdingly, we studied the ihtervarietal variations in the most important floral characteristics for increasing natural outcrossing viz. the number of protruding stigmas per spikelet and pstil length for seed parent, and the number of residual pollren at the anther protruding stage for pollen parent. The materials used in this experiment were Temperate (115 samples of 57 cultivars) and Tropical Japonica (18 samples of 18 cultivars) as shown in Table 1. The Temperate Japonica were classified into 14 improved cultivars and 43 local cultivars. The local cultivars were divided into group A and B. Group A means 16 cultivars which were those without femark rec6rded as "Matjiri", and group B meahs 85 samples qf 27 cultivars which were thbse with th same culrivar name and contained those with remark recorded as "Majiri". "Majiri" means contamintion. Natural outcrossin.g is considered as a cause of the occurente of the contamination.

View full abstract

The variation in culm length of rice plant is infiuenced not only by a gene or genes controlling culm length itself but also by days to heading. These two aspects should be distinguished and seperately analyzed. This paper, consists of two experiments, deals with the variation in culm length analyzed from the standpoints mentioned above. In the first experiment, the data for 93 photosensitive varieties sown at four different times were analyzed. The original data were taken by Dr. H. I. OKA in Taiwan, in 1948-1949. In the second experiment, F₂ plants and 73 randomly selected F₃ lines derived from a cross of a tall Indicaxa Japanese semidwarf variety, Jikkoku were used. Days to heading and culm length were recorded in both experiments. Analysis of covariance in the first experiment showed that the differences of regression coeflicients were not highly significant among sowing times and among varieties. Mean regression coeflieients for four sowing times and for 93 varieties were nearly the same (0.48 and 0.42, respectively). The fact that regression coefficients amonba varieties were not different indicates the deviation of culm length from common regression on days to heading are statistically divided into two sources of variation, regression on days to heading and deviation from the regression.

View full abstract

To determine how many pollen grains are needed for seed set of seed parent of hybrid rice, correlation between the number of pollen grains deposited on the stigma lobes of each spikelet and seed set was observed in male sterile plants which were exposed to air-borne pollen for different periods. A considerable number of spikelets could set seed when only one pollen grain was deposited. We concluded that 3 to 4 pollen grains are practically sufficient for seed set of male sterile line. With regards to flowering duration of spikelet, 30 minutes or more are required to deposit 2 or more pollen grains on the stigma lobes of more than 80% spikelets in the male sterile line used.

View full abstract