1. Experiments have been carried out to determine the relation between the moisture content of artificially dried grass and the relative humidity of the surrounding atmosphere. This relation is given by a smooth curve. At relative humidities which are typical of outdoor winter conditions in the British Isles (80–90%) the equilibrium moisture content lies between 18 and 30%.2. The rate of moulding is directly related to the relative humidity and therefore to the moisture content of the dried grass. For reasonably safe storage a relative humidity of 67 % should not be exceeded. This corresponds to a moisture content of roughly 13 %. Even this low value does not confer absolute immunity from mould growth.3. The importance of extending storage trials over long periods is stressed. In one instance storage for 300 days was required before mould growth occurred. Neglect of this fact may account for the widely held view that a relative humidity of 70–75 % provides safe storage conditions.4. Attempts to detect mould growth at an early stage by plating methods proved unsuccessful. It is shown that the two earliest and most reliable indications of mould growth are (i) the appearance of visible mycelium when the product is examined under a low-power objective, and (ii) the detection of a musty smell.