Although there are varying estimates as to the degree of enzyme hydration required for activity, a threshold value of ca. 0.2 g of water per gram of protein has been widely accepted. The evidence upon which this is based is reviewed here. In particular, results from the use of gas–phase substrates are discussed. Results using solid–phase enzyme–substrate mixtures are not altogether in accord with those obtained using gas–phase substrates. The use of gaseous substrates and products provides an experimental system in which the hydration of the enzyme can be easily controlled, but which is not limited by diffusion. All the results show that increasing hydration enhances activity. The results using gas–phase substrates do not support the existence of a critical hydration value below which enzymatic activity is absent, and suggest that enzyme activity is possible at much lower hydrations than previously thought; they do not support the notion that significant hydration of the surface polar groups is required for activity. However, the marked improvement of activity as hydration is increased suggests that water does play a role, perhaps in optimizing the structure or facilitating the flexibility required for maximal activity.