To evaluate grazing strategies of cattle on pastures, four Holstein black and white coating heifers were Submitted to five sward heights of mombacagrass (Panicum maximum Jacq.) pasture, in a completely randomized design, with two replications, between February and April, 2002. The experimental animals were evaluated during 45 minutes tests. Number of bites, number of feeding stations, number of steps and feeding time were measured, using chronometers and counters. The number of feeding stations per minute decreased quadratically ((Y) over cap = 16.3199 - 0.2424 x + 0.000996 x(2), R-2 = 0.9328, P = 0.0031), from low to tall sward, as a result of the greater available mass present in the highest sward heights, that reduced the searching for other feeding stations. The number of steps between feeding stations increased ((Y) over cap = 0.7247 + 0.01603 x, R-2 = 0.6172, P = 0.0043) increasing the exploration of pasture environment. With the increase in available forage dry matter, the animals reduced total number of steps ((Y) over cap = 513.998287 - 2.659875 x, R-2 = 0.728076, P = 0.0016). steps by minute ((Y) over cap = 11.567445 - 0.058396 x, R-2 = 0.724633, P = 0.0013) and, consequently, the intensity of area utilization ((Y) over cap = 9.533139 - 0.051351 x, R-2 0.7708, P = 0.0005), as the greater sward height assured high forage intake. The animals showed adaptive feeding strategies to different pasture structures, in an attempt to improve their ingestive efficiency.