The effect of an unsaturated water vapor environment on the rate of cement paste hydration depends on the duration of initial hydration under saturated conditions. The minimum duration of curing required to meet performance objectives depends upon the specific performance criterion. When the performance criterion is a rate of hydration equal to that of a continuously saturated specimen, the minimum curing time appears to be 7 d for all w/c values ranging from 0.3 to 0.5. When the criterion is microstructural development, the time required for the paste to develop a sufficiently dense microstructure to mitigate the effects of exposure appears to coincide with the formation of a differential scanning calorimetry (DSC) freezing peak near -30 ºC. For the 0.30 w/c, 0.40 w/c, and 0.50 w/c pastes studied here, the -30 ºC peak forms at 1 d, 3 d, and 14 d, respectively. Arguably, the disappearance of the -30 ºC peak may be a more reliable indicator of microstructural development, but further experiments are needed to justify this. A colloidal microstructural model is used to argue that the appearance of the -30 ºC freezing peak coincides with the time at which the capillary pores are no longer percolated. Consistent with this hypothesis, samples exposed to 90 % RH after the appearance of the -30 ºC freezing peak continued to hydrate at a rate equal to, or greater than, the rate of hydration under sealed conditions. As a result, observation of the DSC freezing peak at -30 ºC in ordinary portland cement concretes is a likely candidate for the development of a field test for the minimum duration of curing required to ensure that subsequent exposure to an unsaturated environment will have little effect on the microstructural development.