Although the microtubular cytoskeleton of plant cells is important in maintaining the direction of cell growth, its natural lability can be harnessed in such a way that new growth axes are permitted. In these circumstances, the system which fabricates the cytoskeleton is presumably responsive to morphogenetic information originating from outside the cell. Spatial patterns of hormonal and metabolic signals within the tissue or organ that house the responsive cells are one possible source of this information. However, a contrasting source takes the form of biophysical information, such as the supracellular patterns of stresses and strains. We examined the microtubular cytoskeleton in roots of tomato and maize to test the assumption that the cortical microtubular array of each cell would have a particular orientation relative to the cell's position within the growth field of the root apex. Accordingly, each intracellular cortical array was mapped to the overall pattern of cells within the apex. In certain areas of the meristem, the arrays seemed to be more variable than elsewhere. These are sites where morphogenetic decisions are taken, usually involving a change in the plane of cell division. Roots which have suffered disturbance to their physical structure (e.g. removal of the root cap), or which had been exposed to low temperatures or treated with certain chemicals (e.g. inhibitors of nuclear division or DNA synthesis), exhibited new patterns of microtubular arrays which in turn predicted novel patterns of cell division. In all these circumstances, the arrays showed consistent alterations within distinct regions of the root - e.g. in the quiescent centre and also in a group of cells just behind the quiescent centre, at the boundary between cortex and stele. These altered arrays indicate that there are supracellular domains in which the microtubules respond to morphogenetic signals. Studies such as these reinforce the concept of microtubule lability and the inherent responsiveness of the microtubular system to external and internal stimuli. However, at present there is no indication of how the morphogenetic programme of the root is set up in the first place. Probably, this is established and stabilized early in embryogenesis and is then perpetuated by the prevailing metabolic and biophysical conditions. The microtubules of the cytoskeleton can be regarded as intracellular automata which not only participate in mitosis and cytokinesis but also ensure the realization of an organogenetic programme. Should the root confront circumstances which temporarily destabilize this programme, the prevailing growth field is sufficiently robust to ensure that the microtubular system is attracted back to the stable, pre-existing state capable of reinstating normal morphogenesis.