Twelve field experiments, carried out over 7 years, examined effects of shading and seed tuber spacing on plant growth, initiation and retention of tubers in four cultivars: Estima, Maris Piper, Maris Peer and Record. Ten of the experiments were carried out at Cambridge and two near Valencia, Spain. Other treatments included in some experiments were floating polythene mulch and planting date.

Shading by up to 75% did not affect the timing of onset or cessation of tuber initiation in Estima but shading by 50% or more delayed the completion of tuber initiation in Maris Piper compared with less severe shading. Except for intensely shaded treatments (50% or more), the majority of tubers were initiated in a very short period (4–7 days). Shading by 37% or more during the period of tuber initiation and increasing planting density, decreased number of tubers per stem initiated in all experiments, but number of tubers was not affected by shading at other stages of growth. At Cambridge, effects of shading on number of tubers >10 mm retained later in growth from normal planting dates (March to early May) were similar to effects on number of tubers initiated, but effects were much reduced or absent following later plantings at Cambridge and in both experiments in Valencia. The decreased effects of shading on number of tubers >10 mm at late plantings at Cambridge were associated with the initiation of fewer tubers at these plantings. Effects of shading, planting density and planting date on number of tubers were a consequence of changes in the frequency of occurrence and tuberization of different stolon types. Increasing shading and planting density and delaying planting reduced the number of lateral and branch stolons and the frequency of their tuberization but there were no effects on number of primary stolons or their tuberization. Consequently, at Cambridge a similar number of tubers was borne on primary stolons in shaded and unshaded crops. In Valencia a greater proportion of initiated tubers was retained at final harvest from shaded treatments than at Cambridge, which accounts for the absence of effects of shading on number of tubers >10 mm. The greater retention of tubers late in growth in Spain may have been associated with the higher peak growth rates achieved in higher radiation fluxes than at Cambridge.

Linear regressions of the data for normal planting dates at Cambridge and from Valencia indicated that the number of tubers >10 mm late in growth was dependent on the radiation environment during the period of tuber initiation. Radiation flux during the brightest period of the first few days of initiation appeared to be the most crucial aspect of radiation affecting number of tubers. As incident radiation can vary greatly over the short period of tuber initiation, it is potentially an important factor affecting number of tubers in field crops.