The impact of various starter phosphorus (P) fertilizers on the growth, nutrient uptake and dry‐matter (DM) yield of forage maize (Zea mais) continuously cropped on the same area and receiving annual, pre‐sowing, broadcast dressings of liquid and semi‐solid dairy manures was investigated in two replicated plot experiments and in whole‐field comparisons in the UK. In Experiment 1 on a shallow calcareous soil (27 mg l−1 Olsen‐extractable P) in 1996, placement of starter P fertilizer (17 or 32 kg ha−1) did not benefit crop growth or significantly (P > 0·05) increase DM yield at harvest. However, in Experiment 2 on a deeper non‐calcareous soil (41 mg l−1 Olsen‐extractable P) in 1997, placement of starter P fertilizer (19 or 41 kg P ha−1), either applied alone or in combination with starter N fertilizer (10 or 25 kg N ha−1), significantly increased early crop growth (P < 0·01) and DM yield at harvest by 1·3 t ha−1 (P < 0·05) compared with a control without starter N or P fertilizer. Placement of starter N fertilizer alone did not benefit early crop growth, but gave similar yields as P, or N and P, fertilizer treatments at harvest. Large treatment differences in N and P uptake by mid‐August had disappeared by harvest. In field comparisons over the 4‐year period 1994–97, the addition of starter P fertilizer increased field cumulative surplus P by over 70%, but without significantly (P > 0·05) increasing DM yield, or nutrient (N and P) uptake, compared with fields that did not receive starter P fertilizer. The results emphasized the extremely low efficiency with which starter P fertilizers are utilized by forage maize and the need to budget manure and fertilizer P inputs more precisely in order to avoid excessive soil P accumulation and the consequent increased risk of P transfer to water causing eutrophication.