We conducted a field experiment to test the hypothesis that improved phosphorus nutrition occurs in maize plants with rapid arbuscular (AM) mycorrhizae development at early developmental stages and that this also is reflected in dry matter allocation and final yield. A split-split plot design was used with previous crop (Zea mays L.-maize and Brassica napus L.-canola), tillage practices (no-tillage or conventional tillage) and P fertilization (5 levels) as factors chosen to modify mycorrhizae development at early developmental stages of maize. Previous cropping with canola resulted in decreased shoot-P concentration and shoot growth of maize at early stages. No-tillage resulted in higher shoot-P concentration but lower shoot weight than conventional tillage. Greater shoot-P uptake was related to a rapid intraradical development of mycorrhizae (previous crop of maize) or rapid connection to a mycorrhizal mycelium network (no-tillage treatments). Maize yield and harvest index were lower after cropping With canola. The yield for conventional tillage was higher than that for no-tillage but the harvest index was lower. The hypothesis was supported at early stages of maize growth by the effect of previous crop but not by results of tillage, because an unknown factor reduced growth in the no-tillage system. The hypothesis was supported at maturity by increased biomass allocation to grain relative to total shoot weight in treatments with greater shoot-P concentration at early stages.