Genetics and development of nine male-sterile tomato mutants

Abstract

Abstract does not appear. First page follows. Summary A search was conducted in commercial plantings of the San Marzano tomato for male-sterile mutants, potentially useful in producing F1 hybrid seed and in cross-breeding. Among 150 unfruitful plants, 12 were found to be genetically male-sterile. Breeding results indicate that the male sterility of each mutant is determined by a single recessive gene. One gene—ms5—was recovered four times. Eight other nonallelic genes—ms6 to ms13—were demonstrated. No mutants for female sterility or cytoplasmic male sterility were found. In the mutant ms5, 3 to 5 per cent of the microsporangia produced functional pollen; by using this pollen in self-pollination, pure-breeding male-sterile populations of ms5 were obtained. No functional pollen was produced by other mutants. The growth of anthers and, to a lesser extent, petals is suppressed to different degrees in the different mutants. The male-sterile segregates in some mutants can be identified macroscopically by the size and form of the floral parts, and in all of the mutants by the anther contents. Anthers are reduced most and breakdown of microsporogenesis occurs earliest in the four mutants whose petal length is significantly diminished. No effects of the genes were observed in other parts of the plant. The genes for male sterility exert their most profound effects on the pollen mother-cells and early microsporogenesis. The mutants differ in time and rate of breakdown. The behavior of chromosomes is normal until the time of breakdown and does not seem to account for failure of pollen development. Each gene affects tapetal development differently; in most mutants the breakdown is delayed, but in a few it occurs at the normal time or earlier. Pollen production is disrupted earliest in those mutants in which tapetal degeneration is delayed. Breakdown of microsporogenesis in ms5 involves an erratic orientation of sporogenous tissue, the latter usurping the locus of the outer tapetum and wall tissue. These effects are determined precisely and consistently, each preparation of a given mutant revealing the same pattern of breakdown. Counts of seed set after cross-pollination and examination of sectioned ovaries revealed no appreciable ovule sterility in any mutant except ms8. In the latter about 25 per cent of the ovules abort—not so many as to disqualify it as a pistillate parent for hybrid-seed production. The rates of natural cross-pollination of certain of the mutants differ with high statistical significance. The highest rate was observed in ms5, which is subject to a low rate of natural self-pollination. In the other eight mutants the rate of natural cross-pollination shows some positive correlation with size of anther tube, suggesting that the insect pollen vectors prefer to visit flowers whose form deviates least from normal. The best mutants for hybrid-seed production are those intermediate in anther abnormality—enough for easy macroscopic identification but not enough to reduce cross-pollination greatly.