Continuous wave nuclear magnetic resonance (NMR) studies indicated that the line width of the water absorption peak (Deltav(1/2)) from crowns of winter and spring wheat (Triticum aestivum L.) increased during cold acclimation. There was a negative correlation between Deltav(1/2) and crown water content, and both of these parameters were correlated with the lowest survival temperature at which 50% or more of the crowns were not killed by freezing (LT(50)). Regression analyses indicated that Deltav(1/2) and water content account for similar variability in LT(50). Slow dehydration of unacclimated winter wheat crowns by artificial means resulted in similarly correlated changes in water content and Deltav(1/2). Rapid dehydration of unacclimated crowns reduced water content but did not influence Deltav(1/2). The incubation of unacclimated winter wheat crowns in a sucrose medium reduced water content and increased Deltav(1/2). The increase in Deltav(1/2) appears to be dependent in part on a reduction in water content and an increase in solutes.Longitudinal (T(1)) and transverse (T(2)) relaxation times of water protons in cereals at different stages of cold acclimation were measured using pulse NMR methods. The T(1) and T(2) signals each demonstrated the existence of two populations of water, one with a short and one with a long relaxation time. During the first 3 weeks of acclimation, the long T(2) decreased significantly in winter-hardy cereals, and did not change in a spring wheat until the 5th week of hardening. There was no change in the long T(1) until the 3rd week of hardening for the winter cereals and until the 7th week of hardening for the spring wheat. No simple relationship could be established between T(1) or T(2) and cold hardiness. Neither continuous wave or pulsed NMR spectroscopy can be used as a diagnostic tool in predicting the cold hardiness of winter wheats. An increase in Deltav(1/2) or a reduction in relaxation times does not provide evidence for ordering of the bulk of the cell water.