Silicon in banana (Musa spp.) : a soil-plant system approach

(eng) Plants can exert a strong imprint on the silicon (Si) continental cycle through their control on the activity of dissolved Si in soils and streams. In plant nutrition, the beneficial effects of silicon on plant yield led to the use of silicate amendments for Si accumulating crops such as rice. Banana is a high nutrient demanding crop belonging to the monocots, which include the majority of Si accumulating plants. Nevertheless, a lot still remains unknown about its Si nutrition. In this PhD thesis, we report on (1) the Si status of banana and (2) the factors influencing this status under controlled and field conditions. At the most common Si concentration in soil solution, Si uptake by banana plants grown in hydroponics is higher than that driven by mass flow,suggesting the existence of active processes in Si uptake. Our data thus provide the first evidence that banana should be classified as a Si accumulator. At the highest Si concentration (1.66 mM), Si uptake is essentially governed by passive transport, suggesting a gradual transition from an accumulating to a non accumulating type. Transpiration of plant organs plays a crucial role in the deposition of Si in the banana plant. Indeed, the distribution of Si among plant organs (roots < pseudostem < leaves) and within one leaf (petiole < midrib < internal lamina < external lamina) clearly follows the transpiration rate of these organs. Studies of banana Si status in hydroponics and under field and greenhouse conditions all reveal that the Si supply largely governs the accumulation of Si in banana. In hydroponics, increasing Si concentration in nutrient solution from 0.08 to 1.66 mM leads to an increase of Si concentration in lamina from 0.9 to 14 g kg-1 DM. The influence of Si supply from soil solution was investigated on two weathering sequences of soils (La Soufrière, Guadeloupe) which strongly differ in their total Si content and their soil constitution. In these soils, the soil-to-plant transfer of Si is directly driven by the concentration of aqueous Si, which is in turn directly governed by the reserve of lithogenic weatherable minerals. Since the banana plant is a Si-accumulator, we believe that the possible positive effects of the Si nutrition on plant resistance to abiotic and biotic stresses should be further studied. Besides, the banana plant appears to be a valuable model plant to study the imprint of vegetation on the Si soil-plant cycle, and thus on the continental and global cycles of Si.