Résumé
Modelling plant growth enables us to predict crop yield, forest harvest and interactions with many ecosystem processes over a long temporal scale. Functional-structural plant models allow us to determine how the allocation of resources affects how a plant occupies space and consequently acquires new resources over time. Although many functional-structural models currently exist to predict plant growth, most models use photosynthesis as the driver for growth. Except for horticultural conditions, this source-driven paradigm lacks effects-based justification. Therefore, we ask: - Are we modelling the correct processes that drive plant growth, and is growth driven mostly by sink (e.g. stems, leaves, roots, inflorescences), or source (photosynthesis) activity? - In current models, is the importance of soil resources (nutrients, water, temperature and their interaction w ith root and shoot meristematic activity) considered adequately? - Do classic models account for architectural adjustment as well as integrating the fundamental principles of development? We put forward the opinion that plant architecture and sink activity must be at the centre of plant grow th models, in particular when the plant grows in limiting conditions. In natural conditions, sinks will more often drive growth than source activity, because sink activity is often controlled by finite soil resources or developmental constraints.