During rainfall events, macropores are generally considered to play a dominant role in infiltration after matrix ponding has occurred. Once ponding has been initiated on the soil matrix, surface runoff may be generated at rainfall intensities less than the saturated hydraulic conductivity of the soil. The amount of runoff will depend on detention storage and how efficiently the surface flow is captured by soil macropores. The efficiency of surface water removal by macropores is diminished if surface vents become clogged sealed by washed-in sediment during the runoff event.Post-event opening of surface vents by the animals that created them can remove evidence of the sealing process and so it is particularly important to examine the temporal stability of the soil surface during rainfall events.In this paper evidence of macropore clogging and post-event clearing of the surface vents is presented. A fine sandy loam passed through a 2 mm diameter sieve was packed into two boxes, each with a surface area of 0.5 m2. The boxes were irrigated at 28 mm h−1 using a low energy rainfall sprinkler. This application rate exceeded the saturated hydraulic conductivity of the soil matrix. After measuring runoff and infiltration from the boxes, one box was held as a control and the second was inoculated with earthworms. After four weeks the inoculated box had a burrow density at the soil surface of 380 m−2, with an average diameter of 5 mm.Macropore sealing occurred immediately after ponding and runoff from the macroporous soil was only 10.7% less than a control with no macropores. Within 24 h after cessation of simulated rainfall the earthworms had cleared washed in material from over 95% of burrow vents. Time to matrix ponding was well predicted using hydraulic parameters characteristic of the soil matrix, indicating that matrix sealing was not significant under the experimental conditions.