The effect of applied cobalt on the cobalt content of pasture plants has been studied in pot, field, and laboratory experiments. The generally poor agreement between pot and field experiments may be partly due to the noted variation of cobalt status of pasture with season and the influence of waterlogging. Cobalt treatments were effective for at least 2 yr on the humus podzol soils, but there was only a residual effect for 1 yr on other soils. Alteration of the soil pH did not prove a satisfactory means of increasing cobalt status of pasture on soils where it was most required. The efficiency of cobalt fertilizers as a means of alleviating problems of cobalt deficiency depends on the manganese content of the soil. Plants grown on soils containing > 1000 p.p.m. of total manganese are unlikely to benefit from application of cobalt fertilizer to the soil, necessitating foliar application by sprays or the use of cobalt oxide pellets for stock.