Study on tuber formation, development and storage in cut flower production system of Gloriosa superba L.

I. Controle of tuber development by controle of node numbers of flower bud differentiation. I-1. Morphological changes in the shoot apex during floral initiation and development. Morphological changes in the shoot apex of Gloriosa superba L. 'Rose Queen' during floral initiation and development were observed using a scanning electron microscope. The sequence of events in floral initiation and development of G. superba was similar to that in Lilium longiflorum except for morphology of perianth primordium in which both inner and outer primordia had the same shape as Tulipa gesneriana. I-2. Genotypic differences in floral initiation and development. Effects of cultivating season on flower bud formation were investigated in three genotypes 'Misato Red', 'Tropical Red' and 'Rose Queen'. Tubers of each genotype were forced to sprout under dry conditions at 30degC for 22-24 days, and then were planted on December 2 (winter culture), on April 3 (spring culture) and on July 3 (summer culture) in a glasshouse, respectively. Most leaves were formed before planting. Floral initiation occurred when about 30 leaves were formed in any season in 'Misato Red' and 'Tropical Red'. In 'Rose Queen', however, floral initiation occurred when 47 leaves had formed in summer culture, and at about 35 leaves in winter and spring culture. The differences in floral initiation among these genotypes seem to be caused by differences in response to temperature, especially high soil temperature which may inhibit floral initiation depending on the genotype. Once floral initiation occurred the flower bud developed rapidly in all culture season. I-3. Effects of soil temperature and temperature to enhance sprouting on node number of flower bud differentiation. Effects of temperature on node numbers of flower bud differentiation of Gloriosa superba L. were investigated using three genotypes, 'Misato Red', 'Tropical Red' and 'Rose Queen'. Tubers of each genotype were kept at 30degC for 23 days to enhance sprouting, and then were planted in a greenhouse with soil temperature controlled at 19.1+-0.6degC or 31.6+-1.0degC. Flower buds initiated at the same node order under both temperature conditions in 'Misato Red' and 'Tropical Red', whereas 'Rose Queen' produced larger numbers of leaves at a higher soil temperature. When tubers were kept at 15 to 40degC for 56 days under dry conditions, flower initiation occurred at 30degC or below in 'Misato Red' and at 35degC or below in 'Tropical Red' during the temperature treatment, but did not occur in 'Rose Queen', indicating a longer juvenile phase of 'Rose Queen'. Whereas 30degC was suitable for leaf differentiation, node numbers of flower bud differentiation were smaller at a lower temperature in all genotypes. 'Misato Red' tubers treated at 30degC for 25 days and 'Tropical Red' and 'Rose Queen' tubers treated for 15 days followed by 15degC for 15 days showed lower node numbers of flower bud differentiation compared with tubers treated at 30degC for 30 days. Since 30degC enhanced vegetative growth but inhibited flower bud differentiation in Gloriosa, tubers kept at 30degC followed by 15degC showed less vegetative growth. The present results show that genotype differences in node number of flower bud differentiation can be due to the differences in juvenile phase length and temperature-related phase transition. I-4. Effect of leaf numbers after cut flower harvest on new tubers development. To clarify the effect of leaf numbers on new tuber development, tubers of 'Misato Red', 'Tropical Red' and 'Rose Queen' were planted on November 10th or May 9th then cut flowers were harvested above 10 to 25th leaf, and finally new tubers were harvested after 60 or 45 days after cut flower harvest, respectively. When tubers were planted in November, 'Misato Red', 'Tropical Red' and 'Rose Queen' required 20, 15 and 25 leaves to produce at least one new tuber as same weigh as mother tuber, indicating the possibility of controlling new tuber weight by the number of left leaves. In the case of May culture, on the other hand, 'Tropical Red' produced equivalent weight of new tubers to mother tubers, but secondary tuber formation was observed regardless number of left leaves. 'Misato Red' and 'Tropical Red' did not produce new tuber with sufficient weight even though 25 leaves were left. In the case of tall genotypes, 'Misato Red' and 'Rose Queen', a larger number of leaves with longer internode in summer lowers workability. Present results shows that the control of new tuber weight by leaf numbers is not practical in summer. Further technical development is needed to prolong culture time without secondary tuber formation.