The paper describes the results of over 300 factorial experiments carried out in each factory area in the years 1934–49 for the Sugar Beet Research and Education Committee of the Ministry of Agriculture as a result of co-operation between Rothamsted workers and the agriculturists and fieldmen attached to the beet factories. All the experiments tested the effects of levels of nitrogen, phosphate and potash, and rather more than half tested the effect of salt also.

Except on fen soils, nitrogen gave substantial responses in all factory areas, especially in the presence of high levels of potash or salt. Large variations in response from season to season were closely associated with the rainfall of the preceding winter months, responses being greater after wet winters than dry ones.

In spite of some selection of sites in favour of greater responses, the average net returns from phosphate were relatively small.

The effect of potash was closely linked with the amount of nitrogen applied; in the presence of nitrogen, dressings well above the level of 1·2 cwt. K2O per acre tested in the experiments are likely to give a useful net return. Soils derived from the Chalky Boulder Clay seem to be exceptional in showing no response. Apart from this, there were only small variations in responses to nitrogen and potash between factory areas. The application of 5 cwt. salt gave substantial responses in almost all parts of the country, whether or not potash was also applied; on the other hand, responses to potash were usually small when salt was also applied.

Whilst there was a general relationship between soil analysis for phosphate and potash (citric acid method) and crop response, adjustments to the optimal dressings according to soil analysis were not of sufficient reliability to be of much practical value.