The type of surface used for running can influence the load that the locomotor apparatus will absorb and the load distribution could be related to the incidence of chronic injuries. As there is no consensus on how the locomotor apparatus adapts to loads originating from running surfaces with different compliance, the objective of this study was to investigate how loads are distributed over the plantar surface while running on natural grass and on a rigid surface--asphalt. Forty-four adult runners with 4+/-3 years of running experience were evaluated while running at 12 km/h for 40 m wearing standardised running shoes and Pedar insoles (Novel). Peak pressure, contact time and contact area were measured in six regions: lateral, central and medial rearfoot, midfoot, lateral and medial forefoot. The surfaces and regions were compared by three ANOVAS (2 x 6). Asphalt and natural grass were statistically different in all variables. Higher peak pressures were observed on asphalt at the central (p<0.001) [grass: 303.8(66.7)kPa; asphalt: 342.3(76.3)kPa] and lateral rearfoot (p<0.001) [grass: 312.7(75.8)kPa; asphalt: 350.9(98.3)kPa] and lateral forefoot (p<0.001) [grass: 221.5(42.9)kPa; asphalt: 245.3(55.5)kPa]. For natural grass, contact time and contact area were significantly greater at the central rearfoot (p<0.001). These results suggest that natural grass may be a surface that provokes lighter loads on the rearfoot and forefoot in recreational runners.