The current view of sulfur nutrition is based on the source‐to‐sink relationship of carbohydrates. SO42‐ reduction is thought to occur mainly in leaves. Surplus reduced sulfur must be transported out of the leaves, loaded into the phloem and transported to other tissues, in particular tissues assumed to be sink organs. However, it has not been proved that tissues which are sinks for carbohydrates are also sink organs for reduced sulfur. It is evident that sinks must communicate with sources, and vice versa, to signal demand and to transport the surplus of reduced sulfur that is produced. The demand‐driven control model of sulfur nutrition proposes that the tripeptide glutathione is the signal which regulates S nutrition of the whole plant at the level of SO42‐ uptake. Acclimatization to environmental changes has been shown to result in several changes in S nutrition of deciduous trees: (i) Drought stress diminished SO42‐ transport into the xylem, although the GSH content in lateral roots remained unaffected, possibly due to an overall reduction in water status. (ii) Flooding decreased APS reductase activity in the anoxic roots. This may be due to enhanced GSH transport to the roots, but it is more likely to be the result of a change in metabolism leading to diminished energy gain in the roots. (iii) Mycorrhization enhanced the GSH content in the phloem, while SO42‐ uptake was not affected. This clearly goes against the demand‐driven control model. (iv) Under both short‐ and long‐term exposure to elevated pCO2, the APS reductase activity in leaves and lateral roots did not correlate with the GSH contents therein. Therefore, it must be assumed that, under these conditions, regulation of S nutrition goes beyond the demand‐driven control model, and occurs within the network of other nutrient metabolism. (v) Atmospheric S in the form of H2S enhanced the reduced sulfur content of the phloem and lateral roots. Under these conditions, the SO42‐ loaded into the xylem decreased. It would appear that the demand‐driven control model of sulfur nutrition is not always valid in the case of deciduous trees.