The dynamics of phosphorus (P) in the soil is of high complexity,due to the low mobility of this nutrient and its reactions with the soil. In order to understand the interactions between phosphorus, soil and plants, the present study was developed to evaluate the efficiency of application modes, combining the application of P to haul with application in tillage. For this, a DBC experiment was set up, with 6 treatments and 4 replications in a first year crop area. The treatments consisted of applying in combination the 120 kg ha-1 P2O5 dose in soybean crop, being 100% under haul; 75% at launch and 25% in line; 50% at launch and 50% in line; 25% haul and 75% in line; 100% in row planting and a witness. Soybean yield, plant height and leaf P content were evaluated. Using the Mehlich-1 and Resin extractors, the behavior of P on the soil profile, its vertical and horizontal mobility were evaluated. In the evaluations of yield, plant height and leaf P content, it was statistically noticed a superiority of the treatments with 50, 75 and 100% of the dose of P applied in planting line. In the evaluation of P present in the soil profile in the depth of 0 - 2.5 cm it was noticed that the treatments with the most of the haul fertilizer (100 and 75%) were statistically superior, and in the depth of 5 - 7.5 cm. Treatments with most of the fertilization in row planting (100 and 75%) are superior. In the evaluation of the vertical and horizontal mobility of P in the soil, it was noticed that there was no movement of this nutrient in the soil, remaining in the application site. The results presented by the extractors, Mehlich 1 and Resin, followed the same tendencies, and the extraction with Resin was more stable, being consistent with the treatments.