Flower color is mainly determined by anthocyanins. Rosa hybrida lacks violet to blue flower varieties due to the absence of delphinidin-based anthocyanins, usually the major constituents of violet and blue flowers, because roses do not possess flavonoid 3',5'-hydoxylase (F3'5'H), a key enzyme for delphinidin biosynthesis. Other factors such as the presence of co-pigments and the vacuolar pH also affect flower color. We analyzed the flavonoid composition of hundreds of rose cultivars and measured the pH of their petal juice in order to select hosts of genetic transformation that would be suitable for the exclusive accumulation of delphinidin and the resulting color change toward blue. Expression of the viola F3'5'H gene in some of the selected cultivars resulted in the accumulation of a high percentage of delphinidin (up to 95%) and a novel bluish flower color. For more exclusive and dominant accumulation of delphinidin irrespective of the hosts, we down-regulated the endogenous dihydroflavonol 4-reductase (DFR) gene and overexpressed the Irisxhollandica DFR gene in addition to the viola F3'5'H gene in a rose cultivar. The resultant roses exclusively accumulated delphinidin in the petals, and the flowers had blue hues not achieved by hybridization breeding. Moreover, the ability for exclusive accumulation of delphinidin was inherited by the next generations.

(fig 1)), ni la croissance des plantes (tab IV), ni le nombre de gousses par plante (tab IV) n'ont été atteints par la contrainte thermique. En

Previous studies demonstrated that there is a final stage and a maximum dry weight for seed abortion in several grain legumes. In this work, the corresponding seed lengths were determined in soybean [Glycine max (L.) Merr.], pea (Pisum sativum L.), and white lupine (Lupinus albus L.). Two genotypes with differing seed size were studied for each species, and the white lupine cultivar Lublanc was placed in two different growing conditions. A photographic technique based on the difference in light transmission between carpels and seeds was used to monitor in a nondestructive manner the lengthening of seeds in their pod. The fate of each seed (aborted vs. nonaborted) was recorded at maturity and time‐courses of seed length were determined from the photos for seeds of the two categories. The lengthening pattern followed by aborted and nonaborted seeds and the maximum lengths attained by each of these categories showed that there was a given length after which seed abortion did not occur. This length was close for the two cultivars of each species. For Lublanc, growing conditions influenced the average size of aborted seeds but did not change the length after which seed abortion did not occur. The respective maximum length of any aborted seed and the length not exceeded by 95% of aborted seeds were 12 and 10 mm for soybean, 8.5 and 6 mm for pea, and 13 and 11.5 mm for white lupine.

Exogenous application of a 2 mol m-3 buffered solution of N6 benzylaminopurine (BAP) to flowers on the main stem inflorescence of Lupinus angustifolius L, cv. Danja profoundly altered reproductive development by reducing post-anthesis abscission of flowers and small pods. The same effect of BAP was recorded for a mutant (abs-) of cv. Danja, in which organ abscission was completely absent, indicating that localized application of cytokinin enhanced reproductive development rather than reduced pedicel abscission per se in the parent line. Application to pedicel and sepals at the open flower stage completely eliminated flower abortion on the main inflorescence, compared with less than 50% pod initiation on untreated inflorescences, more than doubled final pod yield on the main inflorescence and increased the number of mature pods on the whole plant by 33%. A single dose of BAP, to an inflorescence which bore flowers ranging in their stage of development from post-anthesis to immature flower buds, significantly increased the number of pods initiated and at final harvest, measured on a per plant basis. A number of synthetic and naturally occurring cytokinins, including zeatin riboside and dihydrozeatin riboside, were also effective. BAP application induced a longer period of flowering and resulted in a considerably thickened raceme. This was most marked at the distal end which showed enhanced cambial development and secondary vascularization compared with untreated controls. The positive effects of BAP application on pod initiation were not restricted to cv. Danja but were found also for cv. Warrah and cv. Gungurru, both of which have enhanced pod set compared with Danja. Enhanced pod initiation on the main inflorescence generally reduced the number of pods developing on branch inflorescences. Additional application of BAP to flowers on branches, even at the most opportune time and at the most effective site, did not enhance pod initiation and, in some cases, significantly reduced pod set on these branches. The data indicate that it would be very difficult to exploit the positive effect of exogenous cytokinin application on pod set in field crops of lupin. However, selection or genetic engineering of plants with higher levels of endogenous cytokinins in flowers or flower parts at anthesis may provide a means by which to assess the importance of this factor in determining yield stability.

A final stage in seed abortion has been identified by Japanese studies on soybeans (G I .v-c-iiie iiiu.B (L.) Merril). It is a significant stage in the development of the yield components and it may be useful in a pod development scale. To find a simple characterization of this stage, the seeds of 26 samples of aborted plus ripe seeds at harvest were first measured and weighed. Twenty samples were of indeterminate soybean (10 cultivars), 4 of white lupin, Lupinus albus L. (4 cultivars), and 2 of pea, Pisum sativum L. (2 cultivars). The fact that there is no overlap of weight values between aborted seeds and ripe ones indicate that the notion of a final stage in seed abortion is of general validity for the 3 species studied. The value after which the seeds no longer abort may be slightly different between samples of different origins. Measures on green seeds sampled during crop growth were then used to determine reference lengths for green seeds corresponding to reference weights and lengths for dry ones. For the 10 soybean cultivars studied, 11 mm was chosen as reference length. For white lupin and pea, reference lengths were between 6 and 12 mm, and 4 and 5.5 mm respectively. Additional key words : Glycine max, L upinus albus, Pisum sativum, stage of development, pod n Ull/ her, seed number. RÉSUMÉ Caractérisation du stade limite d'avortement des grains chez le soja de type indéterminé, le lupin blanc et le pois.