The effects of different manurial treatments on the yield and mineral composition of early potatoes are described. The crops were grown annually from 1954 to 1964 in a rotation of vegetable crops in a factorial manurial experiment on a sandy loam at Wellesbourne and, in i960 only, on a very fine sandy loam at Efford. At Wellesbourne, farmyard manure (20 tons per acre per crop) with fertilizers consistently produced much higher yields than mineral fertilizers alone, and the relative effect of FYM increased from 1954 to 1959 but not subsequently. In years when moisture supplies were adequate, the larger part of the increase in yield could be attributed to effects of FYM on the early growth of the plants, and a smaller part to the increased rate of bulking of the tubers on the FYM plots. In dry years, however, growth during the harvest period was much slower on the fertilizer plots than on the FYM plots and the additional relative increase in yield on the latter during this period caused the overall effect of FYM to be larger than in the wetter years. In the absence of FYM, each of the major nutrients N, P and K gave significant yield increases, but the need for adequate applications of phosphate and potash fertilizers together was emphasized by a marked interaction between their effects on yield. Applications of phosphate fertilizers had a particularly important effect early in the bulking period, whereas the effects of the nitrogenous fertilizers were greater at the later stages. In the presence of FYM, the fertilizer applications had significant effects only at the later stages of bulking, but at this time the amounts needed to give maximum yields were similar to those needed in the absence of FYM (124 lb. N, 36 lb. P2O5 144 lb. K2O (estimated), each per acre). At Efford, applications of up to 4 cwt. per acre of sulphate of ammonia and 2 cwt. per acre of superphosphate (18% P22O5) gave significant increases in yield, and higher rates would probably have given still higher yields. Potash fertilizers did not affect yields although they markedly increased the potash contents of the plants. The differences between the responses at the two sites are discussed in relation to soil conditions. The leaves of plants manured with FYM (Wellesbourne) contained much more potassium and slightly more phosphorus than leaves from unmanured plants and, while the differences in potassium content were larger towards the end of the bulking period than at the beginning, the reverse was true of the phosphorus contents. Thus, to use leaf composition data to indicate possible shortages of nutrients in the soil, the samples of leaves for chemical analysis must be taken at several different times during the growth of the crop. The value, if any, of extra nutrients for FYM-manured plants cannot at present be assessed. At Efford, applications of sulphate of potash increased the potassium content of the leaves to levels approaching those found in the manured plants at Wellesbourne and higher than those in the unmanured plants. Because of marked interactions between the effects of the phosphate and potash fertilizers at Wellesbourne and indications that phosphate applications at Efford may have been too low to ensure maximum yields, it does not seem possible to combine the data from the two experiments in order to suggest levels of the major nutrients in the potato leaves, at any particular stage or stages, which may be considered adequate for maximal yield production. At Wellesbourne the standard error per plot was smaller for the manured series (8·1%) than for the unmanured (12·9%), indicating that growth was more uniform on the former.